Antigen-antibody binding is a critical process in the immune response. Although recent progress has advanced antibody design, current methods lack a generative framework for end-to-end modeling of full-atom antibody structures and struggle to fully exploit antigen-specific geometric information for optimizing local binding interfaces and global structures. To overcome these limitations, we introduce AbFlow, a paratope-restricted one-step flow-matching framework for designing full-atom antibodies end-to-end. AbFlow incorporates an extended velocity field network featuring an equivariant Surface Multi-channel Encoder, which uses surface-level antigen interaction data to refine the antibody structure, particularly the CDR-H3 region. Extensive experiments in paratope-centric antibody design, multi-CDRs and full-atom antibody design, binding affinity optimization, and complex structure prediction show that AbFlow produces superior antigen-antibody complexes, especially at the contact interface, and markedly improves the binding affinity of generated antibodies.

Equivariant Graph Neural Networks (GNNs) have achieved remarkable success across diverse scientific applications. However, existing approaches face critical efficiency challenges when scaling to large geometric graphs and suffer significant performance degradation when the input graphs are sparsified for computational tractability. To address these limitations, we introduce FastEGNN and DistEGNN, two novel enhancements to equivariant GNNs for large-scale geometric graphs. FastEGNN employs a key innovation: a small ordered set of virtual nodes that effectively approximates the large unordered graph of real nodes. Specifically, we implement distinct message passing and aggregation mechanisms for different virtual nodes to ensure mutual distinctiveness, and minimize Maximum Mean Discrepancy (MMD) between virtual and real coordinates to achieve global distributedness. This design enables FastEGNN to maintain high accuracy while efficiently processing large-scale sparse graphs. For extremely large-scale geometric graphs, we present DistEGNN, a distributed extension where virtual nodes act as global bridges between subgraphs in different devices, maintaining consistency while dramatically reducing memory and computational overhead. We comprehensively evaluate our models across four challenging domains: N-body systems (100 nodes), protein dynamics (800 nodes), Water-3D (8,000 nodes), and our new Fluid113 K benchmark (113,000 nodes). Results demonstrate superior efficiency and performance, establishing new capabilities in large-scale equivariant graph learning.

Understanding and leveraging the 3D structures of proteins is central to a variety of biological and drug discovery tasks. While deep learning has been applied successfully for structure-based protein function prediction tasks, current methods usually employ distinct training for each task. However, each of the tasks is of small size, and such a single-task strategy hinders the models' performance and generalization ability. As some labeled 3D protein datasets are biologically related, combining multi-source datasets for larger-scale multi-task learning is one way to overcome this problem. In this paper, we propose a neural network model to address multiple tasks jointly upon the input of 3D protein structures. In particular, we first construct a standard structure-based multi-task benchmark called Protein-MT, consisting of 6 biologically relevant tasks, including affinity prediction and property prediction, integrated from 4 public datasets. Then, we develop a novel graph neural network for multi-task learning, dubbed Heterogeneous Multichannel Equivariant Network (HeMeNet), which is E(3) equivariant and able to capture heterogeneous relationships between different atoms. Besides, HeMeNet can achieve task-specific learning via the task-aware readout mechanism. Extensive evaluations of our benchmark verify the effectiveness of multi-task learning, and our model generally surpasses state-of-the-art models.

Continual pre-training (CPT) has been an important approach for adapting language models to specific domains or tasks. In this paper, we comprehensively study its key designs to balance the new abilities while retaining the original abilities, and present an effective CPT method that can greatly improve the Chinese language ability and scientific reasoning ability of LLMs. To achieve it, we design specific data mixture and curriculum strategies based on existing datasets and synthetic high-quality data. Concretely, we synthesize multidisciplinary scientific QA pairs based on related web pages to guarantee the data quality, and also devise the performance tracking and data mixture adjustment strategy to ensure the training stability. For the detailed designs, we conduct preliminary studies on a relatively small model, and summarize the findings to help optimize our CPT method. Extensive experiments on a number of evaluation benchmarks show that our approach can largely improve the performance of Llama-3 (8B), including both the general abilities (+8.81 on C-Eval and +6.31 on CMMLU) and the scientific reasoning abilities (+12.00 on MATH and +4.13 on SciEval). Our model, data, and codes are available at https://github.com/RUC-GSAI/Llama-3-SynE.

Yu Sun, Xingyu Qian, Weiwen Xu, Hao Zhang, Chenghao Xiao, Long Li, Deli Zhao, Wenbing Huang, Tingyang Xu, Qifeng Bai, Yu Rong. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Multi-person interactive motion generation, a critical yet under-explored domain in computer character animation, poses significant challenges such as intricate modeling of inter-human interactions beyond individual motions and generating two motions with huge differences from one text condition. Current research often employs separate module branches for individual motions, leading to a loss of interaction information and increased computational demands. To address these challenges, we propose a novel, unified approach that models multi-person motions and their interactions within a single latent space. Our approach streamlines the process by treating interactive motions as an integrated data point, utilizing a Variational AutoEncoder (VAE) for compression into a unified latent space, and performing a diffusion process within this space, guided by the natural language conditions. Experimental results demonstrate our method’s superiority over existing approaches in generation quality, performing text condition in particular when motions have significant asymmetry, and accelerating the generation efficiency while preserving high quality.

Understanding the 3D structure of RNA is essential for deciphering its function and developing RNA-based therapeutics. Geometric Graph Neural Networks (GeoGNNs) that conform to the $\mathrm{E}(3)$-symmetry have advanced RNA structure evaluation, a crucial step toward RNA structure prediction. However, existing GeoGNNs are still defective in two aspects: 1. inefficient or incapable of capturing the full geometries of RNA; 2. limited generalization ability when the size of RNA significantly differs between training and test datasets. In this paper, we propose EquiRNA, a novel equivariant GNN model by exploring the three-level hierarchical geometries of RNA. At its core, EquiRNA effectively addresses the size generalization challenge by reusing the representation of nucleotide, the common building block shared across RNAs of varying sizes. Moreover, by adopting a scalarization-based equivariant GNN as the backbone, our model maintains directional information while offering higher computational efficiency compared to existing GeoGNNs. Additionally, we propose a size-insensitive $K$-nearest neighbor sampling strategy to enhance the model's robustness to RNA size shifts. We test our approach on our created benchmark as well as an existing dataset. The results show that our method significantly outperforms other state-of-the-art methods, providing a robust baseline for RNA 3D structure modeling and evaluation.

Denoising learning of 3D molecules learns molecular representations by imposing noises into the equilibrium conformation and predicting the added noises to recover the equilibrium conformation, which essentially captures the information of molecular force fields. Due to the specificity of Potential Energy Surfaces, the probabilities of physically reasonable noises for each atom in different molecules are different. However, existing methods apply the shared heuristic hand-crafted noise sampling strategy to all molecules, resulting in inaccurate force field learning. In this paper, we propose a novel 3D molecular pre-training method, namely DenoiseVAE, which employs a Noise Generator to acquire atom-specific noise distributions for different molecules. It utilizes the stochastic reparameterization technique to sample noisy conformations from the generated distributions, which are inputted into a Denoising Module for denoising. The Noise Generator and the Denoising Module are jointly learned in a manner conforming with the paradigm of Variational Auto Encoder. Consequently, the sampled noisy conformations can be more diverse, adaptive, and informative, and thus DenoiseVAE can learn representations that better reveal the molecular force fields. Extensive experiments show that DenoiseVAE outperforms the current state-of-the-art methods on various molecular property prediction tasks, demonstrating the effectiveness of it.

Modeling human-like action-to-reaction generation has significant real-world applications, like human-robot interaction and games. Despite recent advancements in single-person motion generation, it is still challenging to well handle action-to-reaction generation, due to the difficulty of directly predicting reaction from action sequence without prompts, and the absence of a unified representation that effectively encodes multi-person motion. To address these challenges, we introduce Think-Then-React (TTR), a large language-model-based framework designed to generate human-like reactions. First, with our fine-grained multimodal training strategy, TTR is capable to unify two processes during inference: a thinking process that explicitly infers action intentions and reasons corresponding reaction description, which serve as semantic prompts, and a reacting process that predicts reactions based on input action and the inferred semantic prompts. Second, to effectively represent multi-person motion in language models, we propose a unified motion tokenizer by decoupling egocentric pose and absolute space features, which effectively represents action and reaction motion with same encoding. Extensive experiments demonstrate that TTR outperforms existing baselines, achieving significant improvements in evaluation metrics, such as reducing FID from 3.988 to 1.942.

Molecular Dynamics (MD) simulations are essential for understanding the atomic-level behavior of molecular systems, giving insights into their transitions and interactions. However, classical MD techniques are limited by the trade-off between accuracy and efficiency, while recent deep learning-based improvements have mostly focused on single-domain molecules, lacking transferability to unfamiliar molecular systems. Therefore, we propose **Uni**fied **Sim**ulator (UniSim), which leverages cross-domain knowledge to enhance the understanding of atomic interactions. First, we employ a multi-head pretraining approach to learn a unified atomic representation model from a large and diverse set of molecular data. Then, based on the stochastic interpolant framework, we learn the state transition patterns over long timesteps from MD trajectories, and introduce a force guidance module for rapidly adapting to different chemical environments. Our experiments demonstrate that UniSim achieves highly competitive performance across small molecules, peptides, and proteins.

Cyclic peptides, characterized by geometric constraints absent in linear peptides, offer enhanced biochemical properties, presenting new opportunities to address unmet medical needs. However, designing target-specific cyclic peptides remains underexplored due to limited training data. To bridge the gap, we propose CP-Composer, a novel generative framework that enables zero-shot cyclic peptide generation via composable geometric constraints. Our approach decomposes complex cyclization patterns into unit constraints, which are incorporated into a diffusion model through geometric conditioning on nodes and edges. During training, the model learns from unit constraints and their random combinations in linear peptides, while at inference, novel constraint combinations required for cyclization are imposed as input. Experiments show that our model, despite trained with linear peptides, is capable of generating diverse target-binding cyclic peptides, reaching success rates from 38% to 84% on different cyclization strategies.

The design of target-specific molecules such as small molecules, peptides, and antibodies is vital for biological research and drug discovery. Existing generative methods are restricted to single-domain molecules, failing to address versatile therapeutic needs or utilize cross-domain transferability to enhance model performance. In this paper, we introduce Unified generative Modeling of 3D Molecules (UniMoMo), the first framework capable of designing binders of multiple molecular domains using a single model. In particular, UniMoMo unifies the representations of different molecules as graphs of blocks, where each block corresponds to either a standard amino acid or a molecular fragment. Based on these unified representations, UniMoMo utilizes a geometric latent diffusion model for 3D molecular generation, featuring an iterative full-atom autoencoder to compress blocks into latent space points, followed by an E(3)-equivariant diffusion process. Extensive benchmarks across peptides, antibodies, and small molecules demonstrate the superiority of our unified framework over existing domain-specific models, highlighting the benefits of multi-domain training.

Accurately predicting 3D structures and dynamics of physical systems is crucial in scientific applications. Existing approaches that rely on geometric Graph Neural Networks (GNNs) effectively enforce $\mathrm{E}(3)$-equivariance, but they often fail in leveraging extensive broader information. While direct application of Large Language Models (LLMs) can incorporate external knowledge, they lack the capability for spatial reasoning with guaranteed equivariance. In this paper, we propose EquiLLM, a novel framework for representing 3D physical systems that seamlessly integrates $\mathrm{E}(3)$-equivariance with LLM capabilities. Specifically, EquiLLM comprises four key components: geometry-aware prompting, an equivariant encoder, an LLM, and an equivariant adapter. Essentially, the LLM guided by the instructive prompt serves as a sophisticated invariant feature processor, while 3D directional information is exclusively handled by the equivariant encoder and adapter modules. Experimental results demonstrate that EquiLLM delivers significant improvements over previous methods across molecular dynamics simulation, human motion simulation, and antibody design, highlighting its promising generalizability.

Lane topology reasoning techniques play a crucial role in high-definition (HD) mapping and autonomous driving applications. While recent years have witnessed significant advances in this field, there has been limited effort to consolidate these works into a comprehensive overview. This survey systematically reviews the evolution and current state of lane topology reasoning methods, categorizing them into three major paradigms: procedural modeling-based methods, aerial imagery-based methods, and onboard sensors-based methods. We analyze the progression from early rule-based approaches to modern learning-based solutions utilizing transformers, graph neural networks (GNNs), and other deep learning architectures. The paper examines standardized evaluation metrics, including road-level measures (APLS and TLTS score), and lane-level metrics (DET and TOP score), along with performance comparisons on benchmark datasets such as OpenLane- V2. We identify key technical challenges, including dataset availability and model efficiency, and outline promising directions for future research. This comprehensive review provides researchers and practitioners with insights into the theoretical frameworks, practical implementations, and emerging trends in lane topology reasoning for HD mapping applications.

Equivariant Graph Neural Networks (GNNs) have demonstrated significant success across various applications. To achieve completeness---that is, the universal approximation property over the space of equivariant functions---the network must effectively capture the intricate multi-body interactions among different nodes. Prior methods attain this via deeper architectures, augmented body orders, or increased degrees of steerable features, often at high computational cost and without polynomial-time solutions. In this work, we present a theoretically grounded framework for constructing complete equivariant GNNs that is both efficient and practical. We prove that a complete equivariant GNN can be achieved through two key components: 1) a complete scalar function, referred to as the canonical form of the geometric graph; and 2) a full-rank steerable basis set. Leveraging this finding, we propose an efficient algorithm for constructing complete equivariant GNNs based on two common models: EGNN and TFN. Empirical results demonstrate that our model demonstrates superior completeness and excellent performance with only a few layers, thereby significantly reducing computational overhead while maintaining strong practical efficacy.

Metal-Organic Frameworks (MOFs) have attracted considerable attention due to their unique properties including high surface area and tunable porosity, and promising applications in catalysis, gas storage, and drug delivery. Structure prediction for MOFs is a challenging task, as these frameworks are intrinsically periodic and hierarchically organized, where the entire structure is assembled from building blocks like metal nodes and organic linkers. To address this, we introduce MOF-BFN, a novel generative model for MOF structure prediction based on Bayesian Flow Networks (BFNs). Given the local geometry of building blocks, MOF-BFN jointly predicts the lattice parameters, as well as the positions and orientations of all building blocks within the unit cell. In particular, the positions are modelled in the fractional coordinate system to naturally incorporate the periodicity. Meanwhile, the orientations are modeled as unit quaternions sampled from learned Bingham distributions via the proposed Bingham BFN, enabling effective orientation generation on the 4D unit hypersphere. Experimental results demonstrate that MOF-BFN achieves state-of-the-art performance across multiple tasks, including structure prediction, geometric property evaluation, and de novo generation, offering a promising tool for designing complex MOF materials.

Accurate prediction of ionic conductivity in electrolyte systems is crucial for advancing numerous scientific and technological applications. While significant progress has been made, current research faces two fundamental challenges: (1) the lack of high-quality standardized benchmarks, and (2) inadequate modeling of geometric structure and intermolecular interactions in mixture systems. To address these limitations, we first reorganize and enhance the CALiSol and DiffMix electrolyte datasets by incorporating geometric graph representations of molecules. We then propose GeoMix, a novel geometry-aware framework that preserves Set-SE(3) equivariance—an essential but challenging property for mixture systems. At the heart of GeoMix lies the Geometric Interaction Network (GIN), an equivariant module specifically designed for intermolecular geometric message passing. Comprehensive experiments demonstrate that GeoMix consistently outperforms diverse baselines (including MLPs, GNNs, and geometric GNNs) across both datasets, validating the importance of cross-molecular geometric interactions and equivariant message passing for accurate property prediction. This work not only establishes new benchmarks for electrolyte research but also provides a general geometric learning framework that advances modeling of mixture systems in energy materials, pharmaceutical development, and beyond.

Coordinate denoising has emerged as a promising method for 3D molecular pretraining due to its theoretical connection to learning molecular force field. However, existing denoising methods rely on oversimplied molecular dynamics that assume atomic motions to be isotropic and homoscedastic. To address these limitations, we propose a novel denoising framework AniDS: Anisotropic Variational Autoencoder for 3D Molecular Denoising. AniDS introduces a structure-aware anisotropic noise generator that can produce atom-specific, full covariance matrices for Gaussian noise distributions to better reflect directional and structural variability in molecular systems. These covariances are derived from pairwise atomic interactions as anisotropic corrections to an isotropic base. Our design ensures that the resulting covariance matrices are symmetric, positive semi-definite, and SO(3)-equivariant, while providing greater capacity to model complex molecular dynamics. Extensive experiments show that AniDS outperforms prior isotropic and homoscedastic denoising models and other leading methods on the MD17 and OC22 benchmarks, achieving average relative improvements of 8.9% and 6.2% in force prediction accuracy. Our case study on a crystal and molecule structure shows that AniDS adaptively suppresses noise along the bonding direction, consistent with physicochemical principles. Our code is available at https://github.com/ZeroKnighting/AniDS.

Designing protein binders targeting specific sites, which requires to generate realistic and functional interaction patterns, is a fundamental challenge in drug discovery. Current structure-based generative models are limited in generating nterfaces with sufficient rationality and interpretability. In this paper, we propose **R**etrieval-**A**ugmented **Di**ffusion for **A**lig**n**ed interfa**ce** (**RADiAnce**), a new framework that leverages known interfaces to guide the design of novel binders. By unifying retrieval and generation in a shared contrastive latent space, our model efficiently identifies relevant interfaces for a given binding site and seamlessly integrates them through a conditional latent diffusion generator, enabling cross-domain interface transfer. Extensive exeriments show that **RADiAnce** significantly outperforms baseline models across multiple metrics, including binding affinity and recovery of geometries and interactions. Additional experimental results validate cross-domain generalization, demonstrating that retrieving interfaces from diverse domains, such as peptides, antibodies, and protein fragments, enhances the generation performance of binders for other domains. Our work establishes a new paradigm for protein binder design that successfully bridges retrieval-based knowledge and generative AI, opening new possibilities for drug discovery.

Abstract Geometric graphs are a special kind of graph with geometric features, which are vital to model many scientific problems. Unlike generic graphs, geometric graphs often exhibit physical symmetries of translations, rotations, and reflections, making them ineffectively processed by current Graph Neural Networks (GNNs). To address this issue, researchers proposed a variety of geometric GNNs equipped with invariant/equivariant properties to better characterize the geometry and topology of geometric graphs. Given the current progress in this field, it is imperative to conduct a comprehensive survey of data structures, models, and applications related to geometric GNNs. In this paper, based on the necessary but concise mathematical preliminaries, we formalize geometric graph as the data structure, on top of which we provide a unified view of existing models from the geometric message passing perspective. Additionally, we summarize the applications as well as the related datasets to facilitate later research for methodology development and experimental evaluation. We also discuss the challenges and future potential directions of geometric GNNs at the end of this survey.

Collaborative filtering (CF) models have demonstrated remarkable performance in recommender systems, which represent users and items as embedding vectors. Recently, due to the powerful modeling capability of graph neural networks for user-item interaction graphs, graph-based CF models have gained increasing attention. They encode each user/item and its subgraph into a single super vector by combining graph embeddings after each graph convolution. However, each hop of the neighbor in the user-item subgraphs carries a specific semantic meaning. Encoding all subgraph information into single vectors and inferring user-item relations with dot products can weaken the semantic information between user and item subgraphs, thus leaving untapped potential. Exploiting this untapped potential provides insight into improving performance for existing recommendation models. To this end, we propose the Graph Cross-correlated Network for Recommendation (GCR), which serves as a general recommendation paradigm that explicitly considers correlations between user/item subgraphs. GCR first introduces the Plain Graph Representation (PGR) to extract information directly from each hop of neighbors into corresponding PGR vectors. Then, GCR develops Cross-Correlated Aggregation (CCA) to construct possible cross-correlated terms between PGR vectors of user/item subgraphs. Finally, GCR comprehensively incorporates the cross-correlated terms for recommendations. Experimental results show that GCR outperforms state-of-the-art models on both interaction prediction and click-through rate prediction tasks.

Hierarchical reinforcement learning (HRL) exhibits remarkable potential in addressing large-scale and long-horizon complex tasks. However, a fundamental challenge, which arises from the inherently entangled nature of hierarchical policies, has not been understood well, consequently compromising the training stability and exploration efficiency of HRL. In this article, we propose a novel HRL algorithm, high-level model approximation (HLMA), presenting both theoretical foundations and practical implementations. In HLMA, a Planner constructs an innovative high-level dynamic model to predict the -step transition of the Controller in a subtask. This allows for the estimation of the evolving performance of the Controller. At low level, we leverage the initial state of each subtask, transforming absolute states into relative deviations by a designed operator as Controller input. This approach facilitates the reuse of subtask domain knowledge, enhancing data efficiency. With this designed structure, we establish the local convergence of each component within HLMA and subsequently derive regret bounds to ensure global convergence. Abundant experiments conducted on complex locomotion and navigation tasks demonstrate that HLMA surpasses other state-of-the-art single-level RL and HRL algorithms in terms of sample efficiency and asymptotic performance. In addition, thorough ablation studies validate the effectiveness of each component of HLMA.

Crystals are the foundation of numerous scientific and industrial applications. While various learning-based approaches have been proposed for crystal generation, existing methods neglect the spacegroup constraint which is crucial in describing the geometry of crystals and closely relevant to many desirable properties. However, considering spacegroup constraint is challenging owing to its diverse and nontrivial forms. In this paper, we reduce the spacegroup constraint into an equivalent formulation that is more tractable to be handcrafted into the generation process. In particular, we translate the spacegroup constraint into two cases: the basis constraint of the invariant exponential space of the lattice matrix and the Wyckoff position constraint of the fractional coordinates. Upon the derived constraints, we then propose DiffCSP++, a novel diffusion model that has enhanced a previous work DiffCSP by further taking spacegroup constraint into account. Experiments on several popular datasets verify the benefit of the involvement of the spacegroup constraint, and show that our DiffCSP++ achieves the best or comparable performance on crystal structure prediction and ab initio crystal generation.

The study of rigid protein-protein docking plays an essential role in a variety of tasks such as drug design and protein engineering. Recently, several learning-based methods have been proposed for the task, exhibiting much faster docking speed than those computational methods. In this paper, we propose a novel learning-based method called ElliDock, which predicts an elliptic paraboloid to represent the protein-protein docking interface. To be specific, our model estimates elliptic paraboloid interfaces for the two input proteins respectively, and obtains the roto-translation transformation for docking by making two interfaces coincide. By its design, ElliDock is independently equivariant with respect to arbitrary rotations/translations of the proteins, which is an indispensable property to ensure the generalization of the docking process. Experimental evaluations show that ElliDock achieves the fastest inference time among all compared methods, and outperforms state-of-the-art learning-based methods, like DiffDock-PP and Alphafold-Multimer, for particularly antibody-antigen docking.

Predicting the binding sites of target proteins plays a fundamental role in drug discovery. Most existing deep-learning methods consider a protein as a 3D image by spatially clustering its atoms into voxels and then feed the voxelized protein into a 3D CNN for prediction. However, the CNN-based methods encounter several critical issues: 1) defective in representing irregular protein structures; 2) sensitive to rotations; 3) insufficient to characterize the protein surface; 4) unaware of protein size shift. To address the above issues, this work proposes EquiPocket, an E(3)-equivariant Graph Neural Network (GNN) for binding site prediction, which comprises three modules: the first one to extract local geometric information for each surface atom, the second one to model both the chemical and spatial structure of protein and the last one to capture the geometry of the surface via equivariant message passing over the surface atoms. We further propose a dense attention output layer to alleviate the effect incurred by variable protein size. Extensive experiments on several representative benchmarks demonstrate the superiority of our framework to the state-of-the-art methods.

Learning policies for multi-entity systems in 3D environments is far more complicated against single-entity scenarios, due to the exponential expansion of the global state space as the number of entities increases. One potential solution of alleviating the exponential complexity is dividing the global space into independent local views that are invariant to transformations including translations and rotations. To this end, this paper proposes Subequivariant Hierarchical Neural Networks (SHNN) to facilitate multi-entity policy learning. In particular, SHNN first dynamically decouples the global space into local entity-level graphs via task assignment. Second, it leverages subequivariant message passing over the local entity-level graphs to devise local reference frames, remarkably compressing the representation redundancy, particularly in gravity-affected environments. Furthermore, to overcome the limitations of existing benchmarks in capturing the subtleties of multi-entity systems under the Euclidean symmetry, we propose the Multi-entity Benchmark (MEBEN), a new suite of environments tailored for exploring a wide range of multi-entity reinforcement learning. Extensive experiments demonstrate significant advancements of SHNN on the proposed benchmarks compared to existing methods. Comprehensive ablations are conducted to verify the indispensability of task assignment and subequivariance.

Many processes in biology and drug discovery involve various 3D interactions between molecules, such as protein and protein, protein and small molecule, etc. Given that different molecules are usually represented in different granularity, existing methods usually encode each type of molecules independently with different models, leaving it defective to learn the various underlying interaction physics. In this paper, we first propose to universally represent an arbitrary 3D complex as a geometric graph of sets, shedding light on encoding all types of molecules with one model. We then propose a Generalist Equivariant Transformer (GET) to effectively capture both domain-specific hierarchies and domain-agnostic interaction physics. To be specific, GET consists of a bilevel attention module, a feed-forward module and a layer normalization module, where each module is E(3) equivariant and specialized for handling sets of variable sizes. Notably, in contrast to conventional pooling-based hierarchical models, our GET is able to retain fine-grained information of all levels. Extensive experiments on the interactions between proteins, small molecules and RNA/DNAs verify the effectiveness and generalization capability of our proposed method across different domains.

In addressing the challenge of Crystal Structure Prediction (CSP), symmetry-aware deep learning models, particularly diffusion models, have been extensively studied, which treat CSP as a conditional generation task. However, ensuring permutation, rotation, and periodic translation equivariance during diffusion process remains incompletely addressed. In this work, we propose EquiCSP, a novel equivariant diffusion-based generative model. We not only address the overlooked issue of lattice permutation equivariance in existing models, but also develop a unique noising algorithm that rigorously maintains periodic translation equivariance throughout both training and inference processes. Our experiments indicate that EquiCSP significantly surpasses existing models in terms of generating accurate structures and demonstrates faster convergence during the training process.

Equivariant Graph Neural Networks (GNNs) have made remarkable success in a variety of scientific applications. However, existing equivariant GNNs encounter the efficiency issue for large geometric graphs and perform poorly if the input is reduced to sparse local graph for speed acceleration. In this paper, we propose FastEGNN, an enhanced model of equivariant GNNs on large geometric graphs. The central idea is leveraging a small ordered set of virtual nodes to approximate the large unordered graph of real nodes. In particular, we distinguish the message passing and aggregation for different virtual node to encourage the mutual distinctiveness, and minimize the Maximum Mean Discrepancy (MMD) between virtual and real coordinates to realize the global distributedness. FastEGNN meets all necessary E(3) symmetries, with certain universal expressivity assurance as well. Our experiments on N-body systems (100 nodes), proteins (800 nodes) and water-3D (8000 nodes), demonstrate that FastEGNN achieves a promising balance between accuracy and efficiency, and outperforms EGNN in accuracy even after dropping all edges in real systems like proteins and water-3D.

Equivariant Graph Neural Networks (GNNs) that incorporate E(3) symmetry have achieved significant success in various scientific applications. As one of the most successful models, EGNN leverages a simple scalarization technique to perform equivariant message passing over only Cartesian vectors (i.e., 1st-degree steerable vectors), enjoying greater efficiency and efficacy compared to equivariant GNNs using higher-degree steerable vectors. This success suggests that higher-degree representations might be unnecessary. In this paper, we disprove this hypothesis by exploring the expressivity of equivariant GNNs on symmetric structures, including $k$-fold rotations and regular polyhedra. We theoretically demonstrate that equivariant GNNs will always degenerate to a zero function if the degree of the output representations is fixed to 1 or other specific values. Based on this theoretical insight, we propose HEGNN, a high-degree version of EGNN to increase the expressivity by incorporating high-degree steerable vectors while maintaining EGNN's efficiency through the scalarization trick. Our extensive experiments demonstrate that HEGNN not only aligns with our theoretical analyses on toy datasets consisting of symmetric structures, but also shows substantial improvements on more complicated datasets such as $N$-body and MD17. Our theoretical findings and empirical results potentially open up new possibilities for the research of equivariant GNNs.

Superconductivity is a fascinating phenomenon observed in certain materials under certain conditions. However, some critical aspects of it, such as the relationship between superconductivity and materials' chemical/structural features, still need to be understood. Recent successes of data-driven approaches in material science strongly inspire researchers to study this relationship with them, but a corresponding dataset is still lacking. Hence, we present a new dataset for data-driven approaches, namely SuperCon3D, containing both 3D crystal structures and experimental superconducting transition temperature (Tc) for the first time. Based on SuperCon3D, we propose two deep learning methods for designing high Tc superconductors. The first is SODNet, a novel equivariant graph attention model for screening known structures, which differs from existing models in incorporating both ordered and disordered geometric content. The second is a diffusion generative model DiffCSP-SC for creating new structures, which enables high Tc-targeted generation. Extensive experiments demonstrate that both our proposed dataset and models are advantageous for designing new high Tc superconducting candidates.

Predicting high-fidelity 3D structures of atomic systems is a fundamental yet challenging problem in scientific domains. While recent work demonstrates the advantage of generative models in this realm, the exploration of different probability paths are still insufficient, and hallucinations during sampling are persistently occurring. To address these pitfalls, we introduce FlowDPO, a novel framework that explores various probability paths with flow matching models and further suppresses hallucinations using Direct Preference Optimization (DPO) for structure generation. Our approach begins with a pre-trained flow matching model to generate multiple candidate structures for each training sample. These structures are then evaluated and ranked based on their distance to the ground truth, resulting in an automatic preference dataset. Using this dataset, we apply DPO to optimize the original model, improving its performance in generating structures closely aligned with the desired reference distribution. As confirmed by our theoretical analysis, such paradigm and objective function are compatible with arbitrary Gaussian paths, exhibiting favorable universality. Extensive experimental results on antibodies and crystals demonstrate substantial benefits of our FlowDPO, highlighting its potential to advance the field of 3D structure prediction with generative models.

Peptide design plays a pivotal role in therapeutics, allowing brand new possibility to leverage target binding sites that are previously undruggable. Most existing methods are either inefficient or only concerned with the target-agnostic design of 1D sequences. In this paper, we propose a generative model for full-atom Peptide design with Geometric LAtent Diffusion (PepGLAD) given the binding site. We first establish a benchmark consisting of both 1D sequences and 3D structures from Protein Data Bank (PDB) and literature for systematic evaluation. We then identify two major challenges of leveraging current diffusion-based models for peptide design: the full-atom geometry and the variable binding geometry. To tackle the first challenge, PepGLAD derives a variational autoencoder that first encodes full-atom residues of variable size into fixed-dimensional latent representations, and then decodes back to the residue space after conducting the diffusion process in the latent space. For the second issue, PepGLAD explores a receptor-specific affine transformation to convert the 3D coordinates into a shared standard space, enabling better generalization ability across different binding shapes. Experimental Results show that our method not only improves diversity and binding affinity significantly in the task of sequence-structure co-design, but also excels at recovering reference structures for binding conformation generation.

Predicting Click-Through Rate (CTR) in billion-scale recommender systems poses a long-standing challenge for Graph Neural Networks (GNNs) due to the overwhelming computational complexity involved in aggregating billions of neighbors. To tackle this, GNN-based CTR models usually sample hundreds of neighbors out of the billions to facilitate efficient online recommendations. However, sampling only a small portion of neighbors results in a severe sampling bias and the failure to encompass the full spectrum of user or item behavioral patterns. To address this challenge, we name the conventional user-item recommendation graph as "micro recommendation grap" and introduce a revolutionizing MAcro Recommendation Graph (MAG) for billion-scale recommendations to reduce the neighbor count from billions to hundreds in the graph structure infrastructure. Specifically, We group micro nodes (users and items) with similar behavior patterns to form macro nodes and then MAG directly describes the relation between the user/item and the hundred of macro nodes rather than the billions of micro nodes. Subsequently, we introduce tailored Macro Graph Neural Networks (MacGNN) to aggregate information on a macro level and revise the embeddings of macro nodes. MacGNN has already served Taobao's homepage feed for two months, providing recommendations for over one billion users. Extensive offline experiments on three public benchmark datasets and an industrial dataset present that MacGNN significantly outperforms twelve CTR baselines while remaining computationally efficient. Besides, online A/B tests confirm MacGNN's superiority in billion-scale recommender systems.

It has been discovered that graph convolutional networks (GCNs) encounter a remarkable drop in performance when multiple layers are piled up. The main factor that accounts for why deep GCNs fail lies in oversmoothing, which isolates the network output from the input with the increase of network depth, weakening expressivity and trainability. In this article, we start by investigating refined measures upon DropEdge-an existing simple yet effective technique to relieve oversmoothing. We term our method as DropEdge++ for its two structure-aware samplers in contrast to DropEdge: layer-dependent (LD) sampler and feature-dependent (FD) sampler. Regarding the LD sampler, we interestingly find that increasingly sampling edges from the bottom layer yields superior performance than the decreasing counterpart as well as DropEdge. We theoretically reveal this phenomenon with mean-edge-number (MEN), a metric closely related to oversmoothing. For the FD sampler, we associate the edge sampling probability with the feature similarity of node pairs and prove that it further correlates the convergence subspace of the output layer with the input features. Extensive experiments on several node classification benchmarks, including both full- and semi-supervised tasks, illustrate the efficacy of DropEdge++ and its compatibility with a variety of backbones by achieving generally better performance over DropEdge and the no-drop version.

Pretraining molecular representation models without labels is fundamental to various applications. Conventional methods mainly process 2D molecular graphs and focus solely on 2D tasks, making their pretrained models incapable of characterizing 3D geometry and thus defective for downstream 3D tasks. In this work, we tackle 3D molecular pretraining in a complete and novel sense. In particular, we first propose to adopt an equivariant energy-based model as the backbone for pretraining, which enjoys the merits of fulfilling the symmetry of 3D space. Then we develop a node-level pretraining loss for force prediction, where we further exploit the Riemann-Gaussian distribution to ensure the loss to be E(3)-invariant, enabling more robustness. Moreover, a graph-level noise scale prediction task is also leveraged to further promote the eventual performance. We evaluate our model pretrained from a large-scale 3D dataset GEOM-QM9 on two challenging 3D benchmarks: MD17 and QM9. Experimental results demonstrate the efficacy of our method against current state-of-the-art pretraining approaches, and verify the validity of our design for each proposed component. Code is available at https://github.com/jiaor17/3D-EMGP.

Molecular perception aims to construct 3D molecules from 3D atom clouds (i.e., atom types and corresponding 3D coordinates), determining bond connections, bond orders, and other molecular attributes within molecules. It is essential for realizing many applications in cheminformatics and bioinformatics, such as modeling quantum chemistry-derived molecular structures in protein-ligand complexes. Additionally, many molecular generation methods can only generate molecular 3D atom clouds, requiring molecular perception as a necessary post-processing. However, existing molecular perception methods mainly rely on predefined chemical rules and fail to leverage 3D geometric information, whose performance is sub-optimal fully. In this study, we propose MPerformer, an SE(3) Transformer-based molecular perceptron exhibiting SE(3)-invariance, to construct 3D molecules from 3D atom clouds efficiently. Besides, we propose a multi-task pretraining-and-finetuning paradigm to learn this model. In the pretraining phase, we jointly minimize an attribute prediction loss and an atom cloud reconstruction loss, mitigating the data imbalance issue of molecular attributes and enhancing the robustness and generalizability of the model. Experiments show that MPerformer significantly outperforms state-of-the-art molecular perception methods in precision and robustness, benefiting various molecular generation scenarios.

Binary Neural Network (BNN) represents convolution weights with 1-bit values, which enhances the efficiency of storage and computation. This paper is motivated by a previously revealed phenomenon that the binary kernels in successful BNNs are nearly power-law distributed: their values are mostly clustered into a small number of codewords. This phenomenon encourages us to compact typical BNNs and obtain further close performance through learning non-repetitive kernels within a binary kernel subspace. Specifically, we regard the binarization process as kernel grouping in terms of a binary codebook, and our task lies in learning to select a smaller subset of codewords from the full code-book. We then leverage the Gumbel-Sinkhorn technique to approximate the codeword selection process, and develop the Permutation Straight-Through Estimator (PSTE) that is able to not only optimize the selection process end-to-end but also maintain the non-repetitive occupancy of selected codewords. Experiments verify that our method reduces both the model size and bit-wise computational costs, and achieves accuracy improvements compared with state-of-the-art BNNs under comparable budgets.

Learning a shared policy that guides the locomotion of different agents is of core interest in Reinforcement Learning (RL), which leads to the study of morphology-agnostic RL. However, existing benchmarks are highly restrictive in the choice of starting point and target point, constraining the movement of the agents within 2D space. In this work, we propose a novel setup for morphology-agnostic RL, dubbed Subequivariant Graph RL in 3D environments (3D-SGRL). Specifically, we first introduce a new set of more practical yet challenging benchmarks in 3D space that allows the agent to have full Degree-of-Freedoms to explore in arbitrary directions starting from arbitrary configurations. Moreover, to optimize the policy over the enlarged state-action space, we propose to inject geometric symmetry, i.e., subequivariance, into the modeling of the policy and Q-function such that the policy can generalize to all directions, improving exploration efficiency. This goal is achieved by a novel SubEquivariant Transformer (SET) that permits expressive message exchange. Finally, we evaluate the proposed method on the proposed benchmarks, where our method consistently and significantly outperforms existing approaches on single-task, multi-task, and zero-shot generalization scenarios. Extensive ablations are also conducted to verify our design.

Antibody design is an essential yet challenging task in various domains like therapeutics and biology. There are two major defects in current learning-based methods: 1) tackling only a certain subtask of the whole antibody design pipeline, making them suboptimal or resource-intensive. 2) omitting either the framework regions or side chains, thus incapable of capturing the full-atom geometry. To address these pitfalls, we propose dynamic Multi-channel Equivariant grAph Network (dyMEAN), an end-to-end full-atom model for E(3)-equivariant antibody design given the epitope and the incomplete sequence of the antibody. Specifically, we first explore structural initialization as a knowledgeable guess of the antibody structure and then propose shadow paratope to bridge the epitope-antibody connections. Both 1D sequences and 3D structures are updated via an adaptive multi-channel equivariant encoder that is able to process protein residues of variable sizes when considering full atoms. Finally, the updated antibody is docked to the epitope via the alignment of the shadow paratope. Experiments on epitope-binding CDR-H3 design, complex structure prediction, and affinity optimization demonstrate the superiority of our end-to-end framework and full-atom modeling.

Assembly Sequence Planning (ASP) is the essential process for modern manufacturing, proven to be NP-complete thus its effective and efficient solution has been a challenge for researchers in the field. In this paper, we present a graph-transformer based framework for the ASP problem which is trained and demonstrated on a self-collected ASP database. The ASP database contains a self-collected set of LEGO models. The LEGO model is abstracted to a heterogeneous graph structure after a thorough analysis of the original structure and feature extraction. The ground truth assembly sequence is first generated by brute-force search and then adjusted manually to be in line with human rational habits. Based on this self-collected ASP dataset, we propose a heterogeneous graph-transformer framework to learn the latent rules for assembly planning. We evaluated the proposed framework in a series of experiments. The results show that the similarity of the predicted and ground truth sequences can reach 0.44, a medium correlation measured by Kendall's τ. Meanwhile, we compared the different effects of node features and edge features and generated a feasible and reasonable assembly sequence as a benchmark for further research. Our dataset and code are available on: htps://github.com/AIR-DISCOVER/ICRA_ASP.

Understanding human interactions within diverse media contexts has emerged as a fundamental challenge. The explosive growth of multimedia data not only provides opportunities for human-centirc analysis but also increases the complexity of processing multimodal data. To address this pivotal challenge and explore its multifaceted dimensions, the Fourth International Workshop on Human-Centric Multimedia Analysis is concentrated on the tasks of human-centric analysis with multimedia and multimodal information. By delving into the nuances of human behavior within multimedia, this workshop aims to uncover novel insights, showcase innovative methodologies, and discuss future directions. With a spotlight on cutting-edge research and a focus on real-world applications, the workshop seeks to equip researchers and practitioners with the tools and knowledge to navigate the intricacies of human-centric multimedia analysis.

Crystal Structure Prediction (CSP) is crucial in various scientific disciplines. While CSP can be addressed by employing currently-prevailing generative models (**e.g.** diffusion models), this task encounters unique challenges owing to the symmetric geometry of crystal structures---the invariance of translation, rotation, and periodicity. To incorporate the above symmetries, this paper proposes DiffCSP, a novel diffusion model to learn the structure distribution from stable crystals. To be specific, DiffCSP jointly generates the lattice and atom coordinates for each crystal by employing a periodic-E(3)-equivariant denoising model, to better model the crystal geometry. Notably, different from related equivariant generative approaches, DiffCSP leverages fractional coordinates other than Cartesian coordinates to represent crystals, remarkably promoting the diffusion and the generation process of atom positions. Extensive experiments verify that our DiffCSP remarkably outperforms existing CSP methods, with a much lower computation cost in contrast to DFT-based methods. Moreover, the superiority of DiffCSP is still observed when it is extended for ab initio crystal generation.

Learning to represent and simulate the dynamics of physical systems is a crucial yet challenging task. Existing equivariant Graph Neural Network (GNN) based methods have encapsulated the symmetry of physics, \emph{e.g.}, translations, rotations, etc, leading to better generalization ability. Nevertheless, their frame-to-frame formulation of the task overlooks the non-Markov property mainly incurred by unobserved dynamics in the environment. In this paper, we reformulate dynamics simulation as a spatio-temporal prediction task, by employing the trajectory in the past period to recover the Non-Markovian interactions. We propose Equivariant Spatio-Temporal Attentive Graph Networks (ESTAG), an equivariant version of spatio-temporal GNNs, to fulfil our purpose. At its core, we design a novel Equivariant Discrete Fourier Transform (EDFT) to extract periodic patterns from the history frames, and then construct an Equivariant Spatial Module (ESM) to accomplish spatial message passing, and an Equivariant Temporal Module (ETM) with the forward attention and equivariant pooling mechanisms to aggregate temporal message. We evaluate our model on three real datasets corresponding to the molecular-, protein- and macro-level. Experimental results verify the effectiveness of ESTAG compared to typical spatio-temporal GNNs and equivariant GNNs.

Visual navigation involves a movable robotic agent striving to reach a point goal (target location) using vision sensory input. While navigation with ideal visibility has seen plenty of success, it becomes challenging in suboptimal visual conditions like poor illumination, where traditional approaches suffer from severe performance degradation. We propose E3VN (echo-enhanced embodied visual navigation) to effectively perceive the surroundings even under poor visibility to mitigate this problem. This is made possible by adopting an echoer that actively perceives the environment via auditory signals. E3VN models the robot agent as playing a cooperative Markov game with that echoer. The action policies of robot and echoer are jointly optimized to maximize the reward in a two-stream actor-critic architecture. During optimization, the reward is also adaptively decomposed into the robot and echoer parts. Our experiments and ablation studies show that E3VN is consistently effective and robust in point goal navigation tasks, especially under nonideal visibility.

The rich content in various real-world networks such as social networks, biological networks, and communication networks provides unprecedented opportunities for unsupervised machine learning on graphs. This paper investigates the fundamental problem of preserving and extracting abundant information from graph-structured data into embedding space without external supervision. To this end, we generalize conventional mutual information computation from vector space to graph domain and present a novel concept, Graphical Mutual Information (GMI), to measure the correlation between input graph and hidden representation. Except for standard GMI which considers graph structures from a local perspective, our further proposed GMI++ additionally captures global topological properties by analyzing the co-occurrence relationship of nodes. GMI and its extension exhibit several benefits: First, they are invariant to the isomorphic transformation of input graphs-an inevitable constraint in many existing methods; Second, they can be efficiently estimated and maximized by current mutual information estimation methods; Lastly, our theoretical analysis confirms their correctness and rationality. With the aid of GMI, we develop an unsupervised embedding model and adapt it to the specific anomaly detection task. Extensive experiments indicate that our GMI methods achieve promising performance in various downstream tasks, such as node classification, link prediction, and anomaly detection.

Multimodal fusion and multitask learning are two vital topics in machine learning. Despite the fruitful progress, existing methods for both problems are still brittle to the same challenge-it remains dilemmatic to integrate the common information across modalities (resp. tasks) meanwhile preserving the specific patterns of each modality (resp. task). Besides, while they are actually closely related to each other, multimodal fusion and multitask learning are rarely explored within the same methodological framework before. In this paper, we propose Channel-Exchanging-Network (CEN) which is self-adaptive, parameter-free, and more importantly, applicable for multimodal and multitask dense image prediction. At its core, CEN adaptively exchanges channels between subnetworks of different modalities. Specifically, the channel exchanging process is self-guided by individual channel importance that is measured by the magnitude of Batch-Normalization (BN) scaling factor during training. For the application of dense image prediction, the validity of CEN is tested by four different scenarios: multimodal fusion, cycle multimodal fusion, multitask learning, and multimodal multitask learning. Extensive experiments on semantic segmentation via RGB-D data and image translation through multi-domain input verify the effectiveness of CEN compared to state-of-the-art methods. Detailed ablation studies have also been carried out, which demonstrate the advantage of each component we propose. Our code is available at https://github.com/yikaiw/CEN.

Sensor nodes have been widely used for mechanical equipment condition monitoring, especially for enclosed metal structures. In these cases, it is important to achieve communication between the sensor nodes inside the enclosed metal structures and the outside receiver node. Conventional wired communication using cables will affect the structural integrity, and wireless communication using electromagnetic waves will be shielded by metal structures, so neither is desirable. Motivated by these issues, this article proposes a Lamb-waves-based sparse distributed penetrating communication (SDPC) system using the propagation characteristics of Lamb waves considered harmful in previous studies to accomplish wireless simultaneous data communication. Combining phase-shift keying and pulse position modulation mechanisms, a phase-position modulation (P-PM) technology is proposed to improve the communication rate. To overcome the intersymbol interference problem caused by the propagation characteristics of Lamb waves, a phase-position sparse reconstruction strategy is proposed for demodulation, which is more convenient and accurate compared with the conventional cross-correlation method. It should be noted that the maximum bit rate without bit error reaches 830.48 kbps via P-PM parameter optimization, which is 8.3 times higher than that ofon–offkeying modulation approach under the same experimental conditions. Finally, distributed communication experiments and simulations further verify the high accuracy, efficiency, and strong robustness of the proposed SDPC system, which can meet the communication requirements of enclosed metal structures.

With the success of the graph embedding model in both academic and industry areas, the robustness of graph embedding against adversarial attack inevitably becomes a crucial problem in graph learning. Existing works usually perform the attack in a white-box fashion: they need to access the predictions/labels to construct their adversarial loss. However, the inaccessibility of predictions/labels makes the white-box attack impractical for a real graph learning system. This paper promotes current frameworks in a more general and flexible sense -- we consider the ability of various types of graph embedding models to remain resilient against black-box driven attacks. We investigate the theoretical connection between graph signal processing and graph embedding models, and formulate the graph embedding model as a general graph signal process with a corresponding graph filter. Therefore, we design a generalized adversarial attack framework: GF-Attack. Without accessing any labels and model predictions, GF-Attack can perform the attack directly on the graph filter in a black-box fashion. We further prove that GF-Attack can perform an effective attack without assumption on the number of layers/window-size of graph embedding models. To validate the generalization of GF-Attack, we construct GF-Attack on five popular graph embedding models. Extensive experiments validate the effectiveness of GF-Attack on several benchmark datasets.

Keypoint detection and description play a central role in computer vision. Most existing methods are in the form of scene-level prediction, without returning the object classes of different keypoints. In this paper, we propose the object-centric formulation, which, beyond the conventional setting, requires further identifying which object each interest point belongs to. With such fine-grained information, our framework enables more downstream potentials, such as object-level matching and pose estimation in a clustered environment. To get around the difficulty of label collection in the real world, we develop a sim2real contrastive learning mechanism that can generalize the model trained in simulation to real-world applications. The novelties of our training method are three-fold: (i) we integrate the uncertainty into the learning framework to improve feature description of hard cases, e.g., less-textured or symmetric patches; (ii) we decouple the object descriptor into two independent branches, intra-object salience and inter-object distinctness, resulting in a better pixel-wise description; (iii) we enforce cross-view semantic consistency for enhanced robustness in representation learning. Comprehensive experiments on image matching and 6D pose estimation verify the encouraging generalization ability of our method. Particularly for 6D pose estimation, our method significantly outperforms typical unsupervised/sim2real methods, achieving a closer gap with the fully supervised counterpart.

Many adaptations of transformers have emerged to address the single-modal vision tasks, where self-attention modules are stacked to handle input sources like images. Intuitively, feeding multiple modalities of data to vision transformers could improve the performance, yet the innermodal attentive weights may be diluted, which could thus greatly undermine the final performance. In this paper, we propose a multimodal token fusion method (TokenFusion), tailored for transformer-based vision tasks. To effectively fuse multiple modalities, TokenFusion dynamically detects uninformative tokens and substitute these tokens with projected and aggregated inter-modal features. Residual positional alignment is also adopted to enable explicit utilization of the inter-modal alignments after fusion. The design of TokenFusion allows the transformer to learn correlations among multimodal features, while the single-modal transformer architecture remains largely intact. Extensive experiments are conducted on a variety of homogeneous and heterogeneous modalities and demonstrate that TokenFusion surpasses state-of-the-art methods in three typical vision tasks: multimodal image-to-image translation, RGB-depth semantic segmentation, and 3D object detection with point cloud and images. Code will be released11https://github.com/huawei-noah/noah-research22https://gitee.com/mindspore/models/tree/master/research/cv/TokenFusion.

3D object detection is a crucial research topic in computer vision, which usually uses 3D point clouds as input in conventional setups. Recently, there is a trend of leveraging multiple sources of input data, such as complementing the 3D point cloud with 2D images that often have richer color and fewer noises. However, due to the heterogeneous geometrics of the 2D and 3D representations, it prevents us from applying off-the-shelf neural networks to achieve multimodal fusion. To that end, we propose Bridged Transformer (BrT), an end-to-end architecture for 3D object detection. BrT is simple and effective, which learns to identify 3D and 2D object bounding boxes from both points and image patches. A key element of BrT lies in the utilization of object queries for bridging 3D and 2D spaces, which unifies different sources of data representations in Transformer. We adopt a form of feature aggregation realized by point-to-patch projections which further strengthen the interaction between images and points. Moreover, BrT works seamlessly for fusing the point cloud with multi-view images. We experimentally show that BrT surpasses state-of-the-art methods on SUN RGB-D and ScanNetV2 datasets.

The Third International Workshop on Human-Centric Multimedia Analysis concentrates on the tasks of human-centric analysis with multimedia and multimodal information. It involves multiple tasks such as face detection and recognition, human body pattern analysis, person re-identification, human action detection, etc. Today, multiple multimedia sensing technologies and large-scale computing infrastructures are emerging at a rapid velocity a wide variety of big multi-modality data for human-centric analysis, which provides rich knowledge to help tackle these challenges. Researchers have strived to push the limits of human-centric multimedia analysis in a wide variety of applications, such as intelligent surveillance, retailing, fashion design, and services. Therefore, this workshop aims to provide a platform to bridge the gap between the communities of human analysis and multimedia.

Getting robots to navigate to multiple objects autonomously is essential yet difficult in robot applications. One of the key challenges is how to explore environments efficiently with camera sensors only. Existing navigation methods mainly focus on fixed cameras and few attempts have been made to navigate with active cameras. As a result, the agent may take a very long time to perceive the environment due to limited camera scope. In contrast, humans typically gain a larger field of view by looking around for a better perception of the environment. How to make robots perceive the environment as efficiently as humans is a fundamental problem in robotics. In this paper, we consider navigating to multiple objects more efficiently with active cameras. Specifically, we cast moving camera to a Markov Decision Process and reformulate the active camera problem as a reinforcement learning problem. However, we have to address two new challenges: 1) how to learn a good camera policy in complex environments and 2) how to coordinate it with the navigation policy. To address these, we carefully design a reward function to encourage the agent to explore more areas by moving camera actively. Moreover, we exploit human experience to infer a rule-based camera action to guide the learning process. Last, to better coordinate two kinds of policies, the camera policy takes navigation actions into account when making camera moving decisions. Experimental results show our camera policy consistently improves the performance of multi-object navigation over four baselines on two datasets.

Graph Neural Networks (GNNs) have become a prevailing tool for learning physical dynamics. However, they still encounter several challenges: 1) Physical laws abide by symmetry, which is a vital inductive bias accounting for model generalization and should be incorporated into the model design. Existing simulators either consider insufficient symmetry, or enforce excessive equivariance in practice when symmetry is partially broken by gravity. 2) Objects in the physical world possess diverse shapes, sizes, and properties, which should be appropriately processed by the model. To tackle these difficulties, we propose a novel backbone, called Subequivariant Graph Neural Network, which 1) relaxes equivariance to subequivariance by considering external fields like gravity, where the universal approximation ability holds theoretically; 2) introduces a new subequivariant object-aware message passing for learning physical interactions between multiple objects of various shapes in particle-based representation; 3) operates in a hierarchical fashion, allowing for modeling long-range and complex interactions. Our model achieves on average over 3% enhancement in contact prediction accuracy across 8 scenarios on Physion and 2$\times$ lower rollout MSE on RigidFall compared with state-of-the-art GNN simulators, while exhibiting strong generalization and data efficiency.

Equivariant Graph neural Networks (EGNs) are powerful in characterizing the dynamics of multi-body physical systems. Existing EGNs conduct flat message passing, which, yet, is unable to capture the spatial/dynamical hierarchy for complex systems particularly, limiting substructure discovery and global information fusion. In this paper, we propose Equivariant Hierarchy-based Graph Networks (EGHNs) which consist of the three key components: generalized Equivariant Matrix Message Passing (EMMP) , E-Pool and E-UnPool. In particular, EMMP is able to improve the expressivity of conventional equivariant message passing, E-Pool assigns the quantities of the low-level nodes into high-level clusters, while E-UnPool leverages the high-level information to update the dynamics of the low-level nodes. As their names imply, both E-Pool and E-UnPool are guaranteed to be equivariant to meet physic symmetry. Considerable experimental evaluations verify the effectiveness of our EGHN on several applications including multi-object dynamics simulation, motion capture, and protein dynamics modeling.

Designing and analyzing model-based RL (MBRL) algorithms with guaranteed monotonic improvement has been challenging, mainly due to the interdependence between policy optimization and model learning. Existing discrepancy bounds generally ignore the impacts of model shifts, and their corresponding algorithms are prone to degrade performance by drastic model updating. In this work, we first propose a novel and general theoretical scheme for a non-decreasing performance guarantee of MBRL. Our follow-up derived bounds reveal the relationship between model shifts and performance improvement. These discoveries encourage us to formulate a constrained lower-bound optimization problem to permit the monotonicity of MBRL. A further example demonstrates that learning models from a dynamically-varying number of explorations benefit the eventual returns. Motivated by these analyses, we design a simple but effective algorithm CMLO (Constrained Model-shift Lower-bound Optimization), by introducing an event-triggered mechanism that flexibly determines when to update the model. Experiments show that CMLO surpasses other state-of-the-art methods and produces a boost when various policy optimization methods are employed.

Molecule generation is central to a variety of applications. Current attention has been paid to approaching the generation task as subgraph prediction and assembling. Nevertheless, these methods usually rely on hand-crafted or external subgraph construction, and the subgraph assembling depends solely on local arrangement. In this paper, we define a novel notion, principal subgraph that is closely related to the informative pattern within molecules. Interestingly, our proposed merge-and-update subgraph extraction method can automatically discover frequent principal subgraphs from the dataset, while previous methods are incapable of. Moreover, we develop a two-step subgraph assembling strategy, which first predicts a set of subgraphs in a sequence-wise manner and then assembles all generated subgraphs globally as the final output molecule. Built upon graph variational auto-encoder, our model is demonstrated to be effective in terms of several evaluation metrics and efficiency, compared with state-of-the-art methods on distribution learning and (constrained) property optimization tasks.

Designing an incentive-compatible auction mechanism that maximizes the auctioneer's revenue while minimizes the bidders’ ex-post regret is an important yet intricate problem in economics. Remarkable progress has been achieved through learning the optimal auction mechanism by neural networks. In this paper, we consider the popular additive valuation and symmetric valuation setting; i.e., the valuation for a set of items is defined as the sum of all items’ valuations in the set, and the valuation distribution is invariant when the bidders and/or the items are permutated. We prove that permutation-equivariant neural networks have significant advantages: the permutation-equivariance decreases the expected ex-post regret, improves the model generalizability, while maintains the expected revenue invariant. This implies that the permutation-equivariance helps approach the theoretically optimal dominant strategy incentive compatible condition, and reduces the required sample complexity for desired generalization. Extensive experiments fully support our theory. To our best knowledge, this is the first work towards understanding the benefits of permutation-equivariance in auction mechanisms.

Detecting 3D keypoints from point clouds is important for shape reconstruction, while this work investigates the dual question: can shape reconstruction benefit 3D keypoint detection? Existing methods either seek salient features according to statistics of different orders or learn to predict keypoints that are invariant to transformation. Nevertheless, the idea of incorporating shape reconstruction into 3D keypoint detection is under-explored. We argue that this is restricted by former problem formulations. To this end, a novel unsupervised paradigm named SNAKE is proposed, which is short for shape-aware neural 3D keypoint field. Similar to recent coordinate-based radiance or distance field, our network takes 3D coordinates as inputs and predicts implicit shape indicators and keypoint saliency simultaneously, thus naturally entangling 3D keypoint detection and shape reconstruction. We achieve superior performance on various public benchmarks, including standalone object datasets ModelNet40, KeypointNet, SMPL meshes and scene-level datasets 3DMatch and Redwood. Intrinsic shape awareness brings several advantages as follows. (1) SNAKE generates 3D keypoints consistent with human semantic annotation, even without such supervision. (2) SNAKE outperforms counterparts in terms of repeatability, especially when the input point clouds are down-sampled. (3) the generated keypoints allow accurate geometric registration, notably in a zero-shot setting. Codes and models are available at https://github.com/zhongcl-thu/SNAKE.

MOTIVATION: The crux of molecular property prediction is to generate meaningful representations of the molecules. One promising route is to exploit the molecular graph structure through graph neural networks (GNNs). Both atoms and bonds significantly affect the chemical properties of a molecule, so an expressive model ought to exploit both node (atom) and edge (bond) information simultaneously. Inspired by this observation, we explore the multi-view modeling with GNN (MVGNN) to form a novel paralleled framework, which considers both atoms and bonds equally important when learning molecular representations. In specific, one view is atom-central and the other view is bond-central, then the two views are circulated via specifically designed components to enable more accurate predictions. To further enhance the expressive power of MVGNN, we propose a cross-dependent message-passing scheme to enhance information communication of different views. The overall framework is termed as CD-MVGNN. RESULTS: We theoretically justify the expressiveness of the proposed model in terms of distinguishing non-isomorphism graphs. Extensive experiments demonstrate that CD-MVGNN achieves remarkably superior performance over the state-of-the-art models on various challenging benchmarks. Meanwhile, visualization results of the node importance are consistent with prior knowledge, which confirms the interpretability power of CD-MVGNN. AVAILABILITY AND IMPLEMENTATION: The code and data underlying this work are available in GitHub at https://github.com/uta-smile/CD-MVGNN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Temporal action localization, which requires a machine to recognize the location as well as the category of action instances in videos, has long been researched in computer vision. The main challenge of temporal action localization lies in that videos are usually long and untrimmed with diverse action contents involved. Existing state-of-the-art action localization methods divide each video into multiple action units (i.e., proposals in two-stage methods and segments in one-stage methods) and then perform action recognition/regression on each of them individually, without explicitly exploiting their relations during learning. In this paper, we claim that the relations between action units play an important role in action localization, and a more powerful action detector should not only capture the local content of each action unit but also allow a wider field of view on the context related to it. To this end, we propose a general graph convolutional module (GCM) that can be easily plugged into existing action localization methods, including two-stage and one-stage paradigms. To be specific, we first construct a graph, where each action unit is represented as a node and their relations between two action units as an edge. Here, we use two types of relations, one for capturing the temporal connections between different action units, and the other one for characterizing their semantic relationship. Particularly for the temporal connections in two-stage methods, we further explore two different kinds of edges, one connecting the overlapping action units and the other one connecting surrounding but disjointed units. Upon the graph we built, we then apply graph convolutional networks (GCNs) to model the relations among different action units, which is able to learn more informative representations to enhance action localization. Experimental results show that our GCM consistently improves the performance of existing action localization methods, including two-stage methods (e.g., CBR [15] and R-C3D [47]) and one-stage methods (e.g., D-SSAD [22]), verifying the generality and effectiveness of our GCM. Moreover, with the aid of GCM, our approach significantly outperforms the state-of-the-art on THUMOS14 (50.9 percent versus 42.8 percent). Augmentation experiments on ActivityNet also verify the efficacy of modeling the relationships between action units. The source code and the pre-trained models are available at https://github.com/Alvin-Zeng/GCM.

We propose a deep fine-grained multi-level fusion architecture for monocular 3D object detection, with an additionally designed anti-occlusion optimization process. Conventional monocular 3D object detection methods usually leverage geometry constraints such as keypoints, object shape relationships, and 3D to 2D optimizations to offset the lack of accurate depth information. However, these methods still struggle against directly extracting rich information for fusion from the depth estimation. To solve the problem, we integrate the monocular 3D features with the pseudo-LiDAR filter generation network between fine-grained multi-level layers. Our network utilizes the inherent multi-scale and promotes depth and semantic information flow in different stages. The new architecture can obtain features that incorporate more reliable depth information. At the same time, the problem of occlusion among objects is prevalent in natural scenes yet remains unsolved mainly. We propose a novel loss function that aims at alleviating the problem of occlusion. Extensive experiments have proved that the framework demonstrates a competitive performance, especially for the complex scenes with occlusion.

Variants of Graph Neural Networks (GNNs) for representation learning have been proposed recently and achieved fruitful results in various fields. Among them, Graph Attention Network (GAT) first employs a self-attention strategy to learn attention weights for each edge in the spatial domain. However, learning the attentions over edges can only focus on the local information of graphs and greatly increases the computational costs. In this paper, we first introduce the attention mechanism in the spectral domain of graphs and present Spectral Graph Attention Network (SpGAT) that learns representations for different frequency components regarding weighted filters and graph wavelets bases. In this way, SpGAT can better capture global patterns of graphs in an efficient manner with much fewer learned parameters than that of GAT. Further, to reduce the computational cost of SpGAT brought by the eigen-decomposition, we propose a fast approximation variant SpGAT-Cheby. We thoroughly evaluate the performance of SpGAT and SpGAT-Cheby in semi-supervised node classification tasks and verify the effectiveness of the learned attentions in the spectral domain.

It has been a challenge to learning skills for an agent from long-horizon unannotated demonstrations. Existing approaches like Hierarchical Imitation Learning(HIL) are prone to compounding errors or suboptimal solutions. In this paper, we propose Option-GAIL, a novel method to learn skills at long horizon. The key idea of Option-GAIL is modeling the task hierarchy by options and train the policy via generative adversarial optimization. In particular, we propose an Expectation-Maximization(EM)-style algorithm: an E-step that samples the options of expert conditioned on the current learned policy, and an M-step that updates the low- and high-level policies of agent simultaneously to minimize the newly proposed option-occupancy measurement between the expert and the agent. We theoretically prove the convergence of the proposed algorithm. Experiments show that Option-GAIL outperforms other counterparts consistently across a variety of tasks.

The Second International Workshop on Human-centric Multimedia Analysis is focused on human-centric analysis using multimedia information. The human-centric multimedia analysis is one of the fundamental and challenging problems of multimedia understanding. It involves various human-centric analysis tasks like face recognition, human pose estimation, person re-identification, human action recognition, person tracking, human-computer interaction, etc. Nowadays, various multimedia sensing devices and large-scale computing infrastructures are generating a wide variety of multi-modality data at a rapid velocity, which supplies rich knowledge to tackle these challenges for human-centric analysis. Researchers and engineers have strived to push the limits of human-centric multimedia analysis in a wide variety of applications, such as smart city, retailing, intelligent manufacturing, and public services. To this end, our workshop aims to provide a platform to promote exchanges and integration for the fields of human analysis and multimedia.

Tactile sensing plays an important role in robotic perception and manipulation tasks. To overcome the real-world limitations of data collection, simulating tactile response in a virtual environment comes as a desirable direction of robotic research. In this paper, we propose Elastic Interaction of Particles (EIP) for tactile simulation, which is capable of reflecting the elastic property of the tactile sensor as well as characterizing the fine-grained physical interaction during contact. Specifically, EIP models the tactile sensor as a group of coordinated particles, and the elastic property is applied to regulate the deformation of particles during contact. With the tactile simulation by EIP, we further propose a tactile-visual perception network that enables information fusion between tactile data and visual images. The perception network is based on a global-to-local fusion mechanism where multi-scale tactile features are aggregated to the corresponding local region of the visual modality with the guidance of tactile positions and directions. The fusion method exhibits superiority regarding the 3D geometric reconstruction task. Our code for EIP is available at https://github.com/yikaiw/EIP.

Social media has been developing rapidly in public due to its nature of spreading new information, which leads to rumors being circulated. Meanwhile, detecting rumors from such massive information in social media is becoming an arduous challenge. Therefore, some deep learning methods are applied to discover rumors through the way they spread, such as Recursive Neural Network (RvNN) and so on. However, these deep learning methods only take into account the patterns of deep propagation but ignore the structures of wide dispersion in rumor detection. Actually, propagation and dispersion are two crucial characteristics of rumors. In this paper, we propose a novel bi-directional graph model, named Bi-Directional Graph Convolutional Networks (Bi-GCN), to explore both characteristics by operating on both top-down and bottom-up propagation of rumors. It leverages a GCN with a top-down directed graph of rumor spreading to learn the patterns of rumor propagation; and a GCN with an opposite directed graph of rumor diffusion to capture the structures of rumor dispersion. Moreover, the information from source post is involved in each layer of GCN to enhance the influences from the roots of rumors. Encouraging empirical results on several benchmarks confirm the superiority of the proposed method over the state-of-the-art approaches.

With the great success of graph embedding model on both academic and industry area, the robustness of graph embedding against adversarial attack inevitably becomes a central problem in graph learning domain. Regardless of the fruitful progress, most of the current works perform the attack in a white-box fashion: they need to access the model predictions and labels to construct their adversarial loss. However, the inaccessibility of model predictions in real systems makes the white-box attack impractical to real graph learning system. This paper promotes current frameworks in a more general and flexible sense – we demand to attack various kinds of graph embedding model with black-box driven. To this end, we begin by investigating the theoretical connections between graph signal processing and graph embedding models in a principled way and formulate the graph embedding model as a general graph signal process with corresponding graph filter. As such, a generalized adversarial attacker: GF-Attack is constructed by the graph filter and feature matrix. Instead of accessing any knowledge of the target classifiers used in graph embedding, GF-Attack performs the attack only on the graph filter in a black-box attack fashion. To validate the generalization of GF-Attack, we construct the attacker on four popular graph embedding models. Extensive experimental results validate the effectiveness of our attacker on several benchmark datasets. Particularly by using our attack, even small graph perturbations like one-edge flip is able to consistently make a strong attack in performance to different graph embedding models.

In this paper, we study Reinforcement Learning from Demonstrations (RLfD) that improves the exploration efficiency of Reinforcement Learning (RL) by providing expert demonstrations. Most of existing RLfD methods require demonstrations to be perfect and sufficient, which yet is unrealistic to meet in practice. To work on imperfect demonstrations, we first define an imperfect expert setting for RLfD in a formal way, and then point out that previous methods suffer from two issues in terms of optimality and convergence, respectively. Upon the theoretical findings we have derived, we tackle these two issues by regarding the expert guidance as a soft constraint on regulating the policy exploration of the agent, which eventually leads to a constrained optimization problem. We further demonstrate that such problem is able to be addressed efficiently by performing a local linear search on its dual form. Considerable empirical evaluations on a comprehensive collection of benchmarks indicate our method attains consistent improvement over other RLfD counterparts.

The technique of recursive neighborhood aggregation has dominated the implementation of existing successful Graph Neural Networks (GNNs). However, the recursive information propagation across layers inevitably brings in extra calculations, potentially large variance, and difficulty of parallel computation. In this paper, we propose Graph Unfolding Networks (GUNets) as an alternative mechanism of recursive neighborhood aggregation for graph representation learning. Comparing to generic GNNs, our proposed GUNets are efficient, robust and practically effective. At their core, GUNets unfold the local structure of every node, i.e. the rooted tree, to a set of trajectories, and then adopt set function to capture the topology of the rooted subtree, which is more convenient for parallel computation than the recursive neighborhood aggregation process. More importantly, through a specific design of the set function, our architecture enables efficient and robust learning on large-scale graphs without resorting to any pruning of the rooted subtree that is usually necessary in generic GNNs. Extensive experiments on five large datasets (the number of nodes ranges from 104 to 106) show that our GUNets achieve comparable or even better results than current successful GNNs while gaining significantly more efficiency and lower accuracy variance. Codes can be found at github.com/GUNets/GUNets.

Recent advances in Graph Convolutional Networks (GCNs) have led to state-of-the-art performance on various graph-related tasks. However, most existing GCN models do not explicitly identify whether all the aggregated neighbors are valuable to the learning tasks, which may harm the learning performance. In this paper, we consider the problem of node classification and propose the Label-Aware Graph Convolutional Network (LAGCN) framework which can directly identify valuable neighbors to enhance the performance of existing GCN models. Our contribution is three-fold. First, we propose a label-aware edge classifier that can filter distracting neighbors and add valuable neighbors for each node to refine the original graph into a label-aware (LA) graph. Existing GCN models can directly learn from the LA graph to improve the performance without changing their model architectures. Second, we introduce the concept of positive ratio to evaluate the density of valuable neighbors in the LA graph. Theoretical analysis reveals that using the edge classifier to increase the positive ratio can improve the learning performance of existing GCN models. Third, we conduct extensive node classification experiments on benchmark datasets. The results verify that LAGCN can improve the performance of existing GCN models considerably, in terms of node classification.

Unsupervised image-to-image translation is a central task in computer vision. Current translation frameworks will abandon the discriminator once the training process is completed. This paper contends a novel role of the discriminator by reusing it for encoding the images of the target domain. The proposed architecture, termed as NICE-GAN, exhibits two advantageous patterns over previous approaches: First, it is more compact since no independent encoding component is required; Second, this plug-in encoder is directly trained by the adversary loss, making it more informative and trained more effectively if a multi-scale discriminator is applied. The main issue in NICE-GAN is the coupling of translation with discrimination along the encoder, which could incur training inconsistency when we play the min-max game via GAN. To tackle this issue, we develop a decoupled training strategy by which the encoder is only trained when maximizing the adversary loss while keeping frozen otherwise. Extensive experiments on four popular benchmarks demonstrate the superior performance of NICE-GAN over state-of-the-art methods in terms of FID, KID, and also human preference. Comprehensive ablation studies are also carried out to isolate the validity of each proposed component. Our codes are available at https://github.com/alpc91/NICE-GAN-pytorch.

We address the problem of video grounding from natural language queries. The key challenge in this task is that one training video might only contain a few annotated starting/ending frames that can be used as positive examples for model training. Most conventional approaches directly train a binary classifier using such imbalance data, thus achieving inferior results. The key idea of this paper is to use the distances between the frame within the ground truth and the starting (ending) frame as dense supervisions to improve the video grounding accuracy. Specifically, we design a novel dense regression network (DRN) to regress the distances from each frame to the starting (ending) frame of the video segment described by the query. We also propose a simple but effective IoU regression head module to explicitly consider the localization quality of the grounding results (i.e., the IoU between the predicted location and the ground truth). Experimental results show that our approach significantly outperforms state-of-the-arts on three datasets (i.e., Charades-STA, ActivityNet-Captions, and TACoS).

Many real data come in the form of non-grid objects, i.e. graphs, from social networks to molecules. Adaptation of deep learning from grid-alike data (e.g. images) to graphs has recently received unprecedented attention from both machine learning and data mining communities, leading to a new cross-domain field---Deep Graph Learning (DGL). Instead of painstaking feature engineering, DGL aims to learn informative representations of graphs in an end-to-end manner. It has exhibited remarkable success in various tasks, such as node/graph classification, link prediction, etc.

The First International Workshop on Human-Centric MultimediaAnalysis is concentrated on the tasks of human-centric analysis with multimedia and multimodal information. It is one of the fundamental and challenging problems of multimedia understanding. The human-centric multimedia analysis involves multiple tasks such as face detection and recognition, human body pattern analysis, person re-identification, human action detection, person tracking,human-object interaction, and so on. Today, multiple multimedia sensing technologies and large-scale computing infrastructures are producing at a rapid velocity a wide variety of big multi-modality data for human-centric analysis, which provides rich knowledge to help tackle these challenges. Researchers have strived to push the limits of human-centric multimedia analysis in a wide variety of applications, such as intelligent surveillance, retailing, fashion design, and services. Therefore, this workshop aims to provide a platform to bridge the gap between the communities of human analysis and multimedia.

How to obtain informative representations of molecules is a crucial prerequisite in AI-driven drug design and discovery. Recent researches abstract molecules as graphs and employ Graph Neural Networks (GNNs) for molecular representation learning. Nevertheless, two issues impede the usage of GNNs in real scenarios: (1) insufficient labeled molecules for supervised training; (2) poor generalization capability to new-synthesized molecules. To address them both, we propose a novel framework, GROVER, which stands for Graph Representation frOm self-superVised mEssage passing tRansformer. With carefully designed self-supervised tasks in node-, edge- and graph-level, GROVER can learn rich structural and semantic information of molecules from enormous unlabelled molecular data. Rather, to encode such complex information, GROVER integrates Message Passing Networks into the Transformer-style architecture to deliver a class of more expressive encoders of molecules. The flexibility of GROVER allows it to be trained efficiently on large-scale molecular dataset without requiring any supervision, thus being immunized to the two issues mentioned above. We pre-train GROVER with 100 million parameters on 10 million unlabelled molecules---the biggest GNN and the largest training dataset in molecular representation learning. We then leverage the pre-trained GROVER for molecular property prediction followed by task-specific fine-tuning, where we observe a huge improvement (more than 6% on average) from current state-of-the-art methods on 11 challenging benchmarks. The insights we gained are that well-designed self-supervision losses and largely-expressive pre-trained models enjoy the significant potential on performance boosting.

Deep multimodal fusion by using multiple sources of data for classification or regression has exhibited a clear advantage over the unimodal counterpart on various applications. Yet, current methods including aggregation-based and alignment-based fusion are still inadequate in balancing the trade-off between inter-modal fusion and intra-modal processing, incurring a bottleneck of performance improvement. To this end, this paper proposes Channel-Exchanging-Network (CEN), a parameter-free multimodal fusion framework that dynamically exchanges channels between sub-networks of different modalities. Specifically, the channel exchanging process is self-guided by individual channel importance that is measured by the magnitude of Batch-Normalization (BN) scaling factor during training. The validity of such exchanging process is also guaranteed by sharing convolutional filters yet keeping separate BN layers across modalities, which, as an add-on benefit, allows our multimodal architecture to be almost as compact as a unimodal network. Extensive experiments on semantic segmentation via RGB-D data and image translation through multi-domain input verify the effectiveness of our CEN compared to current state-of-the-art methods. Detailed ablation studies have also been carried out, which provably affirm the advantage of each component we propose. Our code is available at https://github.com/yikaiw/CEN.

The richness in the content of various information networks such as social networks and communication networks provides the unprecedented potential for learning high-quality expressive representations without external supervision. This paper investigates how to preserve and extract the abundant information from graph-structured data into embedding space in an unsupervised manner. To this end, we propose a novel concept, Graphical Mutual Information (GMI), to measure the correlation between input graphs and high-level hidden representations. GMI generalizes the idea of conventional mutual information computations from vector space to the graph domain where measuring mutual information from two aspects of node features and topological structure is indispensable. GMI exhibits several benefits: First, it is invariant to the isomorphic transformation of input graphs—an inevitable constraint in many existing graph representation learning algorithms; Besides, it can be efficiently estimated and maximized by current mutual information estimation methods such as MINE; Finally, our theoretical analysis confirms its correctness and rationality. With the aid of GMI, we develop an unsupervised learning model trained by maximizing GMI between the input and output of a graph neural encoder. Considerable experiments on transductive as well as inductive node classification and link prediction demonstrate that our method outperforms state-of-the-art unsupervised counterparts, and even sometimes exceeds the performance of supervised ones.

Temporal action localization aims to accurately localize and recognize all possible action instances from an untrimmed video automatically. Most existing methods perform this task by first generating a set of proposals and then recognizing each independently. However, due to the complex structures and large content variations in action instances, recognizing them individually can be difficult. Fortunately, some proposals often share information regarding one specific action. Such information, which is ignored in existing methods, can be used to boost recognition performance. In this paper, we propose a novel mechanism, called relation attention, to exploit informative relations among proposals based on their appearance or optical flow features. Specifically, we propose a relation attention module to enhance representation power by capturing useful information from other proposals. This module does not change the dimensions of the original input and output and does not rely on any specific proposal generation methods or feature extraction backbone networks. Experimental results show that the proposed relation attention mechanism improves performance significantly on both Thumos14 and ActivityNet1.3 datasets compared to existing architectures. For example, relying on Structured Segment Networks (SSN), the proposed relation attention module helps to increase the mAP from 41.4 to 43.7 on the Thumos14 dataset and outperforms the state-of-the-art results.

Tactile recognition aims at identifying target objects according to tactile sensory readings. Tactile data have two salient properties: 1) sequentially real-time and 2) temporally correlated, which essentially calls for a real-time (i.e., online fixed-budget) and recurrent recognition procedure. Based on an efficient and robust spatio-temporal feature representation for tactile sequences, we handle the problem of real-time recurrent tactile recognition by proposing a bounded online-sequential learning framework, and incorporates the strength of batch-regularization bootstrapping, bounded recursive reservoir, and momentum-based estimation. Experimental evaluations show that it outperforms the state-of-the-art methods by a large margin on test accuracy; and its training performance is superior to most compared models from aspects of average online training error, computational complexity, and storage efficiency.

The recent advances in deep learning have made it possible to generate photo-realistic images by using neural networks and even to extrapolate video frames from an input video clip. In this paper, for the sake of both furthering this exploration and our own interest in a realistic application, we study imageto-video translation and particularly focus on the videos of facial expressions. This problem challenges the deep neural networks by another temporal dimension comparing to the image-to-image translation. Moreover, its single input image fails most existing video generation methods that rely on recurrent models. We propose a user-controllable approach so as to generate video clips of various lengths from a single face image. The lengths and types of the expressions are controlled by users. To this end, we design a novel neural network architecture that can incorporate the user input into its skip connections and propose several improvements to the adversarial training method for the neural network. Experiments and user studies verify the effectiveness of our approach. Especially, we would like to highlight that even for the face images in the wild (downloaded from the Web and the authors’ own photos), our model can generate high-quality facial expression videos of which about 50% are labeled as real by Amazon Mechanical Turk workers.

The goal of task transfer in reinforcement learning is migrating the action policy of an agent to the target task from the source task. Given their successes on robotic action planning, current methods mostly rely on two requirements: exactlyrelevant expert demonstrations or the explicitly-coded cost function on target task, both of which, however, are inconvenient to obtain in practice. In this paper, we relax these two strong conditions by developing a novel task transfer framework where the expert preference is applied as a guidance. In particular, we alternate the following two steps: Firstly, letting experts apply pre-defined preference rules to select related expert demonstrates for the target task. Secondly, based on the selection result, we learn the target cost function and trajectory distribution simultaneously via enhanced Adversarial MaxEnt IRL and generate more trajectories by the learned target distribution for the next preference selection. The theoretical analysis on the distribution learning and convergence of the proposed algorithm are provided. Extensive simulations on several benchmarks have been conducted for further verifying the effectiveness of the proposed method.

Most of the recent progresses on visual question answering are based on recurrent neural networks (RNNs) with attention. Despite the success, these models are often timeconsuming and having difficulties in modeling long range dependencies due to the sequential nature of RNNs. We propose a new architecture, Positional Self-Attention with Coattention (PSAC), which does not require RNNs for video question answering. Specifically, inspired by the success of self-attention in machine translation task, we propose a Positional Self-Attention to calculate the response at each position by attending to all positions within the same sequence, and then add representations of absolute positions. Therefore, PSAC can exploit the global dependencies of question and temporal information in the video, and make the process of question and video encoding executed in parallel. Furthermore, in addition to attending to the video features relevant to the given questions (i.e., video attention), we utilize the co-attention mechanism by simultaneously modeling “what words to listen to” (question attention). To the best of our knowledge, this is the first work of replacing RNNs with selfattention for the task of visual question answering. Experimental results of four tasks on the benchmark dataset show that our model significantly outperforms the state-of-the-art on three tasks and attains comparable result on the Count task. Our model requires less computation time and achieves better performance compared with the RNNs-based methods. Additional ablation study demonstrates the effect of each component of our proposed model.

Unsupervised domain adaptation (UDA) transfers knowledge from a label-rich source domain to a fully-unlabeled target domain. To tackle this task, recent approaches resort to discriminative domain transfer in virtue of pseudo-labels to enforce the class-level distribution alignment across the source and target domains. These methods, however, are vulnerable to the error accumulation and thus incapable of preserving cross-domain category consistency, as the pseudo-labeling accuracy is not guaranteed explicitly. In this paper, we propose the Progressive Feature Alignment Network (PFAN) to align the discriminative features across domains progressively and effectively, via exploiting the intra-class variation in the target domain. To be specific, we first develop an Easy-to-Hard Transfer Strategy (EHTS) and an Adaptive Prototype Alignment (APA) step to train our model iteratively and alternatively. Moreover, upon observing that a good domain adaptation usually requires a non-saturated source classifier, we consider a simple yet efficient way to retard the convergence speed of the source classification loss by further involving a temperature variate into the soft-max function. The extensive experimental results reveal that the proposed PFAN exceeds the state-of-the-art performance on three UDA datasets.

We propose a simple, fast, and accurate one-stage approach to visual grounding, inspired by the following insight. The performances of existing propose-and-rank two-stage methods are capped by the quality of the region candidates they propose in the first stage - if none of the candidates could cover the ground truth region, there is no hope in the second stage to rank the right region to the top. To avoid this caveat, we propose a one-stage model that enables end-to-end joint optimization. The main idea is as straightforward as fusing a text query's embedding into the YOLOv3 object detector, augmented by spatial features so as to account for spatial mentions in the query. Despite being simple, this one-stage approach shows great potential in terms of both accuracy and speed for both phrase localization and referring expression comprehension, according to our experiments. Given these results along with careful investigations into some popular region proposals, we advocate for visual grounding a paradigm shift from the conventional two-stage methods to the one-stage framework.

Most state-of-the-art action localization systems process each action proposal individually, without explicitly exploiting their relations during learning. However, the relations between proposals actually play an important role in action localization, since a meaningful action always consists of multiple proposals in a video. In this paper, we propose to exploit the proposal-proposal relations using GraphConvolutional Networks (GCNs). First, we construct an action proposal graph, where each proposal is represented as a node and their relations between two proposals as an edge. Here, we use two types of relations, one for capturing the context information for each proposal and the other one for characterizing the correlations between distinct actions. Then we apply the GCNs over the graph to model the relations among different proposals and learn powerful representations for the action classification and localization. Experimental results show that our approach significantly outperforms the state-of-the-art on THUMOS14(49.1% versus 42.8%). Moreover, augmentation experiments on ActivityNet also verify the efficacy of modeling action proposal relationships.

We introduce the Neural Collaborative Subspace Clustering, a neural model that discovers clusters of data points drawn from a union of low-dimensional subspaces. In contrast to previous attempts, our model runs without the aid of spectral clustering. This makes our algorithm one of the kinds that can gracefully scale to large datasets. At its heart, our neural model benefits from a classifier which determines whether a pair of points lies on the same subspace or not. Essential to our model is the construction of two affinity matrices, one from the classifier and the other from a notion of subspace self-expressiveness, to supervise training in a collaborative scheme. We thoroughly assess and contrast the performance of our model against various state-of-the-art clustering algorithms including deep subspace-based ones.

Humans have a surprising capacity to induce general rules that describe the specific actions portrayed in a video sequence. The rules learned through this kind of process allow us to achieve similar goals to those shown in the video but in more general circumstances. Enabling an agent to achieve the same capacity represents a significant challenge. In this paper, we propose a Watch-Reason-Code(WRC) model to synthesise programs that describe the process carried out in a set of video sequences. The 'watch' stage is simply a video encoder that encodes videos to multiple feature vectors. The 'reason' stage takes as input the features from multiple diverse videos and generates a compact feature representation via a novel deviation-pooling method. The 'code' stage is a multi-sound decoder that the first step leverages to generate a draft program layout with possible useful statements and perceptions. Further steps then take these outputs and generate a fully structured, compile-able and executable program. We evaluate the effectiveness of our model in two video-to-program synthesis environments, Karel andVizDoom, showing that we can achieve the state-of-the-art under a variety of settings.

There has been a prominent emergence of work on video prediction, aiming to extrapolate the future video frames from the past. Existing temporal-based methods are limited to certain numbers of frames. In this paper, we study video prediction from a single still image in the facial expression domain, a.k.a, facial image-to-video translation. Our main approach, dubbed AffineGAN, associates each facial image with an expression intensity and leverages an affine transformation in the latent space. AffineGAN allows users to control the number of frames to predict as well as the expression intensity for each of them. Unlike previous intensity-based methods, We derive an inverse formulation to the affine transformation, enabling automatic inference of the facial expression intensities from videos --- manual annotation is not only tedious but also ambiguous as people express in various ways and have different opinions about the intensity of a facial image. Both quantitative and qualitative results verify the superiority of AffineGAN over the state of the arts. Notably, in a Turing test with web faces, more than 50% of the facial expression videos generated by AffineGAN are considered real by the Amazon Mechanical Turk workers. This work could improve users' communication experience by enabling them to conveniently and creatively produce expression GIFs, which are popular art forms in online messaging and social networks.

This paper studies Learning from Observations (LfO) for imitation learning with access to state-only demonstrations. In contrast to Learning from Demonstration (LfD) that involves both action and state supervisions, LfO is more practical in leveraging previously inapplicable resources (e.g., videos), yet more challenging due to the incomplete expert guidance. In this paper, we investigate LfO and its difference with LfD in both theoretical and practical perspectives. We first prove that the gap between LfD and LfO actually lies in the disagreement of inverse dynamics models between the imitator and expert, if following the modeling approach of GAIL. More importantly, the upper bound of this gap is revealed by a negative causal entropy which can be minimized in a model-free way. We term our method as Inverse-Dynamics-Disagreement-Minimization (IDDM) which enhances the conventional LfO method through further bridging the gap to LfD. Considerable empirical results on challenging benchmarks indicate that our method attains consistent improvements over other LfO counterparts.

Node classification and graph classification are two graph learning problems that predict the class label of a node and the class label of a graph respectively. A node of a graph usually represents a real-world entity, e.g., a user in a social network, or a protein in a protein-protein interaction network. In this work, we consider a more challenging but practically useful setting, in which a node itself is a graph instance. This leads to a hierarchical graph perspective which arises in many domains such as social network, biological network and document collection. For example, in a social network, a group of people with shared interests forms a user group, whereas a number of user groups are interconnected via interactions or common members. We study the node classification problem in the hierarchical graph where a “node” is a graph instance, e.g., a user group in the above example. As labels are usually limited in real-world data, we design two novel semi-supervised solutions named SEmi-supervised grAph cLassification via Cautious/Active Iteration (or SEAL-C/AI in short). SEAL-C/AI adopt an iterative framework that takes turns to build or update two classifiers, one working at the graph instance level and the other at the hierarchical graph level. To simplify the representation of the hierarchical graph, we propose a novel supervised, self-attentive graph embedding method called SAGE, which embeds graph instances of arbitrary size into fixed-length vectors. Through experiments on synthetic data and Tencent QQ group data, we demonstrate that SEAL-C/AI not only outperform competing methods by a significant margin in terms of accuracy/Macro-F1, but also generate meaningful interpretations of the learned representations.

Tactile sensing is becoming an indispensable robotic ability for object recognition and grasping manipulation despite dealing with tactile data as the force distribution over the array sensors continuously changes as a function of time. In this paper, we propose an efficient feature extractor named linear dynamic systems based fuzzy C-means method (LDS) to encode the tactile sequences, both spatially and temporally. To this end, we decompose every input sequence into multiple subsequences, each of which is locally described by a finite-ordered observability matrix of the LDS model. A fuzzy c-means method is then applied to cluster the local LDS descriptors for learning a codebook. Conditioned on the resulting codebook, the global tactile representation is formulated by employing two different frameworks to integrate the subsequences within each tactile sequence, namely, the Vector of locally aggregated descriptor and Bag-of-Word approaches. The effectiveness of the proposed model is verified by a variety of experimental evaluations on five benchmark datasets. Results reveal that our proposed method achieves a higher classification accuracy than the state-of-the-art models with a large margin.

In this paper, we propose the novel principal backpropagation networks (PBNets) to revisit the backpropagation algorithms commonly used in training two-stream networks for video action recognition. We content that existing approaches always take all the frames/snippets for the backpropagation not optimal for video recognition since the desired actions only occur in a short period within a video. To remedy these drawbacks, we design a watch-and-choose mechanism. In particular, the watching stage exploits a dense snippet-wise temporal pooling strategy to discover the global characteristic for each input video, while the choosing phase only backpropagates a small number of representative snippets that are selected with two novel strategies, i.e., Max-rule and KL-rule. We prove that with the proposed selection strategies, performing the backpropagation on the selected subset is capable of decreasing the loss of the whole snippets as well. The proposed PBNets are evaluated on two standard video action recognition benchmarks UCF101 and HMDB51, where it surpasses the state of the arts consistently, but requiring less memory and computation to achieve high performance.

Taking the feature pyramids into account has become a crucial way to boost the object detection performance. While various pyramid representations have been developed, previous works are still inefficient to integrate the semantical information over different scales. Moreover, recent object detectors are suffering from accurate object location applications, mainly due to the coarse definition of the "positive" examples at training and predicting phases. In this paper, we begin by analyzing current pyramid solutions, and then propose a novel architecture by reconfiguring the feature hierarchy in a flexible yet effective way. In particular, our architecture consists of two lightweight and trainable processes: global attention and local reconfiguration. The global attention is to emphasize the global information of each feature scale, while the local reconfiguration is to capture the local correlations across different scales. Both the global attention and local reconfiguration are non-linear and thus exhibit more expressive ability. Then, we discover that the loss function for object detectors during training is the central cause of the inaccurate location problem. We propose to address this issue by reshaping the standard cross entropy loss such that it focuses more on accurate predictions. Both the feature reconfiguration and the consistent loss could be utilized in popular one-stage (SSD, RetinaNet) and two-stage (Faster R-CNN) detection frameworks. Extensive experimental evaluations on PASCAL VOC 2007, PASCAL VOC 2012 and MS COCO datasets demonstrate that, our models achieve consistent and significant boosts compared with other state-of-the-art methods.

We address the challenging problem of weakly supervised temporal action localization from unconstrained web videos, where only the video-level action labels are available during training. Inspired by the adversarial erasing strategy in weakly supervised semantic segmentation, we propose a novel iterative-winners-out network. Specifically, we make two technical contributions: we propose an iterative training strategy, namely, winners-out, to select the most discriminative action instances in each training iteration and remove them in the next training iteration. This iterative process alleviates the "winner-takes-all" phenomenon that existing approaches tend to choose the video segments that strongly correspond to the video label but neglects other less discriminative video segments. With this strategy, our network is able to localize not only the most discriminative instances but also the less discriminative ones. To better select the target action instances in winners-out, we devise a class-discriminative localization technique. By employing the attention mechanism and the information learned from data, our technique is able to identify the most discriminative action instances effectively. The two key components are integrated into an end-to-end network to localize actions without using the frame-level annotations. Extensive experimental results demonstrate that our method outperforms the state-of-the-art weakly supervised approaches on ActivityNet1.3 and improves mAP from 16.9% to 20.5% on THUMOS14. Notably, even with weak video-level supervision, our method attains comparable accuracy to those employing frame-level supervisions.

Stably grasping objects for a specific task is a hot research topic in robotics due to multiple degrees of freedom of hand kinematics, various shapes of objects, and incomplete visual sensing of objects (partial point clouds). This paper proposes an effective grasp planning method by integrating the crucial grasp cues (positions and orientations of thumb fingertips and the wrist) from humans’ grasp experience. This approach has multiple advantages: greatly reducing the search space of the hand kinematics; no reconstruction or registration; being able to directly perform on the partial point cloud of objects. Meanwhile, for various shapes of objects which are partially observable in the single-view visual sensing, the presented approach learns the “thumb” grasp point employing a signature of histograms of orientations shape descriptor based on objects’ category level. This method recognizes the grasp point according to the shape affordance at each point on the object, which performs the grasp point generalization on the familiar objects. Finally, we verify the developed methods via both simulations and experiments by grasping various shapes of objects.

Despite the recent success of end-to-end learned representations, hand-crafted optical flow features are still widely used in video analysis tasks. To fill this gap, we propose TVNet, a novel end-to-end trainable neural network, to learn optical-flow-like features from data. TVNet subsumes a specific optical flow solver, the TV-L1 method, and is initialized by unfolding its optimization iterations as neural layers. TVNet can therefore be used directly without any extra learning. Moreover, it can be naturally concatenated with other task-specific networks to formulate an end-to-end architecture, thus making our method more efficient than current multi-stage approaches by avoiding the need to pre-compute and store features on disk. Finally, the parameters of the TVNet can be further fine-tuned by end-to-end training. This enables TVNet to learn richer and task-specific patterns beyond exact optical flow. Extensive experiments on two action recognition benchmarks verify the effectiveness of the proposed approach. Our TVNet achieves better accuracies than all compared methods, while being competitive with the fastest counterpart in terms of features extraction time.

Dense event captioning aims to detect and describe all events of interest contained in a video. Despite the advanced development in this area, existing methods tackle this task by making use of dense temporal annotations, which is dramatically source-consuming. This paper formulates a new problem: weakly supervised dense event captioning, which does not require temporal segment annotations for model training. Our solution is based on the one-to-one correspondence assumption, each caption describes one temporal segment, and each temporal segment has one caption, which holds in current benchmark datasets and most real world cases. We decompose the problem into a pair of dual problems: event captioning and sentence localization and present a cycle system to train our model. Extensive experimental results are provided to demonstrate the ability of our model on both dense event captioning and sentence localization in videos.

Graph Convolutional Networks (GCNs) have become a crucial tool on learning representations of graph vertices. The main challenge of adapting GCNs on large-scale graphs is the scalability issue that it incurs heavy cost both in computation and memory due to the uncontrollable neighborhood expansion across layers. In this paper, we accelerate the training of GCNs through developing an adaptive layer-wise sampling method. By constructing the network layer by layer in a top-down passway, we sample the lower layer conditioned on the top one, where the sampled neighborhoods are shared by different parent nodes and the over expansion is avoided owing to the fixed-size sampling. More importantly, the proposed sampler is adaptive and applicable for explicit variance reduction, which in turn enhances the training of our method. Furthermore, we propose a novel and economical approach to promote the message passing over distant nodes by applying skip connections. Intensive experiments on several benchmarks verify the effectiveness of our method regarding the classification accuracy while enjoying faster convergence speed.

Linear Dynamical Systems (LDSs) are fundamental tools for modeling spatio-temporal data in various disciplines. Though rich in modeling, analyzing LDSs is not free of difficulty, mainly because LDSs do not comply with Euclidean geometry and hence conventional learning techniques can not be applied directly. In this paper, we propose an efficient projected gradient descent method to minimize a general form of a loss function and demonstrate how clustering and sparse coding with LDSs can be solved by the proposed method efficiently. To this end, we first derive a novel canonical form for representing the parameters of an LDS, and then show how gradient-descent updates through the projection on the space of LDSs can be achieved dexterously. In contrast to previous studies, our solution avoids any approximation in LDS modeling or during the optimization process. Extensive experiments reveal the superior performance of the proposed method in terms of the convergence and classification accuracy over state-of-the-art techniques.

Robotic tactile recognition aims at identifying target objects or environments from tactile sensory readings. The advancement of unsupervised feature learning and biological tactile sensing inspire us proposing the model of 3T-RTCN that performs spatio-temporal feature representation and fusion for tactile recognition. It decomposes tactile data into spatial and temporal threads, and incorporates the strength of randomized tiling convolutional networks. Experimental evaluations show that it outperforms some state-of-the-art methods with a large margin regarding recognition accuracy, robustness, and fault-tolerance; we also achieve an order-of-magnitude speedup over equivalent networks with pretraining and finetuning. Practical suggestions and hints are summarized in the end for effectively handling the tactile data.

Linear Dynamical Systems (LDSs) are the fundamental tools for encoding spatio-temporal data in various disciplines. To enhance the performance of LDSs, in this paper, we address the challenging issue of performing sparse coding on the space of LDSs, where both data and dictionary atoms are LDSs. Rather than approximate the extended observability with a finite-order matrix, we represent the space of LDSs by an infinite Grassmannian consisting of the orthonormalized extended observability subspaces. Via a homeomorphic mapping, such Grassmannian is embedded into the space of symmetric matrices, where a tractable objective function can be derived for sparse coding. Then, we propose an efficient method to learn the system parameters of the dictionary atoms explicitly, by imposing the symmetric constraint to the transition matrices of the data and dictionary systems. Moreover, we combine the state covariance into the algorithm formulation, thus further promoting the performance of the models with symmetric transition matrices. Comparative experimental evaluations reveal the superior performance of proposed methods on various tasks including video classification and tactile recognition.

In this paper, we propose a novel categorization framework to recognize tactile sequences based on two particular properties of the tactile data. For the first one, tactile sequences are spatio-temporal data which is sequential and dynamic, depicting the process of grasping an object in different grasping stages; therefore, it is reasonable to discover the dynamical pattern by modeling tactile data as integral sequences rather than individual frames. For the second one, a tactile sequence contains various dynamical patterns in different stages of the grasping process; therefore, we decompose the whole sequence into multiple mini-sequences so as to enhance feature resolution. To address both properties in our framework, we take advantage of a Bag-of-System model using parameters of the Linear Dynamic System (LDS) as feature descriptors. Moreover, we employ the LDS with Symmetric Transition matrix (LDSST) rather than the original LDS as the building-block in order to obtain accurate codewords of the codebook of the Bag-of-System. The performance of our framework is evaluated on six real-world databases of three groups. Our experiments show that classification using LDSST is better than the original LDS, and the decomposition of tactile sequences does improve the accuracy of classification. The experiment results also show the superiority of our framework in comparison with other state-of-the-art sequence classifiers.

The perception of the tactile and vision modalities plays a crucial role in object classification and robot precise operations. On one hand, the visual perception helps acquire the objects apparent characteristics (e.g., shape and color) that are paramount for grasp planning. On the other hand, when grasping the object, the tactile sensors can detect the softness or stiffness and the surface texture of the object, thus can be used to classify the objects. In this paper, two different tactile models are first developed to model the tactile sequences, namely bag-of-system (BoS) and deep dynamical system (DDS). To be specific, BoS applies the extreme learning machine method as the classifier, and DDS employs deep neural networks as the feature mapping functions in order to deal with the highly nonlinear data. As for visual sensing, we formulate a signature of histograms of orientations descriptor to model the shape of objects. Moreover, by using human experience, we identify the graspable components of objects on the category level instead of the object level. Finally, a fast planning method considering both contact points exaction and hand kinematics is proposed to accomplish the grasping manipulation. Comparative experiments demonstrate the effectiveness of our proposed methods on the tasks including the robot grasping and the object classification.

The optimization method based extreme learning machine (optimization-based ELM) is generalized from single-hidden-layer feed-forward neural networks (SLFNs) by making use of kernels instead of neuron-alike hidden nodes. This approach is known for its high scalability, low computational complexity, and mild optimization constrains. The multi-kernel learning (MKL) framework Simple MKL iteratively determines the combination of kernels by gradient descent wrapping a standard support vector machine (SVM) solver. Simple MKL can be applied to many kinds of supervised learning problems to receive a more stable performance with rapid convergence speed. This paper proposes a new approach: MK-ELM (multi-kernel extreme learning machine) that applies Simple MKL framework to the optimization-based ELM algorithm. The performance analysis on binary classification problems with various scales shows that MK-ELM tends to achieve the best generalization performance as well as being the most insensitive to parameters comparing to optimization-based ELM and Simple MKL. As a result, MK-ELM can be implemented in real applications easily.