One global model in federated learning (FL) might not be sufficient to serve many clients with non-IID tasks and distributions. Despite recent advances in FL to train multiple global models for better personalization, they only provide limited model choices to clients, so local finetuning of multiple models is still indispensable. This paper proposes a novel ``FedMerge'' approach that can create a single personalized model per client by simply merging multiple global models with automatically optimized and customized weights. We formulate this problem as a joint optimization of global models and the merging weights per client. Unlike existing FL approaches, where the server broadcasts one or multiple global models to all clients, the server only needs to send a customized, merged model to each client. Moreover, instead of periodically interrupting the local training and re-initializing it to a global model, the merged model aligns better with each client's task and data distribution, smoothening the local-global gap between consecutive rounds caused by client drift. We evaluate FedMerge on different non-IID settings applied to various domains with diverse tasks and data types, in which FedMerge consistently outperforms existing FL approaches, including clustering-based and mixture-of-experts (MoE) based methods.

Federated foundation models represent a new paradigm to jointly fine-tune pre-trained foundation models across clients. It is still a challenge to fine-tune foundation models for a small group of new users or specialized scenarios, which typically involve limited data compared to the large-scale data used in pre-training. In this context, the trade-off between personalization and federation becomes more sensitive. To tackle these, we proposed a bi-level personalization framework for federated fine-tuning on foundation models. Specifically, we conduct personalized fine-tuning on the client-level using its private data, and then conduct a personalized aggregation on the server-level using similar users measured by client-specific task vectors. Given the personalization information gained from client-level fine-tuning, the server-level personalized aggregation can gain group-wise personalization information while mitigating the disturbance of irrelevant or interest-conflict clients with non-IID data. The effectiveness of the proposed algorithm has been demonstrated by extensive experimental analysis in benchmark datasets.

Training a general-purpose time series foundation models with robust generalization capabilities across diverse applications from scratch is still an open challenge. Efforts are primarily focused on fusing cross-domain time series datasets to extract shared subsequences as tokens for training models on Transformer architecture. However, due to significant statistical heterogeneity across domains, this cross-domain fusing approach doesn't work effectively as the same as fusing texts and images. To tackle this challenge, this paper proposes a novel federated learning approach to address the heterogeneity in time series foundation models training, namely FFTS. Specifically, each data-holding organization is treated as an independent client in a collaborative learning framework with federated settings, and then many client-specific local models will be trained to preserve the unique characteristics per dataset. Moreover, a new regularization mechanism will be applied to both client-side and server-side, thus to align the shared knowledge across heterogeneous datasets from different domains. Extensive experiments on benchmark datasets demonstrate the effectiveness of the proposed federated learning approach. The newly learned time series foundation models achieve superior generalization capabilities on cross-domain time series analysis tasks, including forecasting, imputation, and anomaly detection.

Federated Collaborative Filtering (FedCF) is an emerging field focused on developing a new recommendation framework with preserving privacy in a federated setting. Existing FedCF methods typically combine distributed Collaborative Filtering (CF) algorithms with privacy-preserving mechanisms, and then preserve personalized information into a user embedding vector. However, the user embedding is usually insufficient to preserve the rich information of the fine-grained personalization across heterogeneous clients. This paper proposes a novel personalized FedCF method by preserving users' personalized information into a latent variable and a neural model simultaneously. Specifically, we decompose the modeling of user knowledge into two encoders, each designed to capture shared knowledge and personalized knowledge separately. A personalized gating network is then applied to balance personalization and generalization between the global and local encoders. Moreover, to effectively train the proposed framework, we model the CF problem as a specialized Variational AutoEncoder (VAE) task by integrating user interaction vector reconstruction with missing value prediction. The decoder is trained to reconstruct the implicit feedback from items the user has interacted with, while also predicting items the user might be interested in but has not yet interacted with. Experimental results on benchmark datasets demonstrate that the proposed method outperforms other baseline methods, showcasing superior performance.

The Base-New Trade-off (BNT) problem universally exists during the optimization of CLIP-based prompt tuning, where continuous fine-tuning on base (target) classes leads to a simultaneous decrease of generalization ability on new (unseen) classes. Existing approaches attempt to regulate the prompt tuning process to balance BNT by appending constraints. However, imposed on the same target prompt, these constraints fail to fully avert the mutual exclusivity between the optimization directions for base and new. As a novel solution to this challenge, we propose the plug-andplay Dual-Prompt Collaboration (DPC) framework, the first that decoupling the optimization processes of base and new tasks at the prompt level. Specifically, we clone a learnable parallel prompt based on the backbone prompt, and introduce a variable Weighting-Decoupling framework to independently control the optimization directions of dual prompts specific to base or new tasks, thus avoiding the conflict in generalization. Meanwhile, we propose a Dynamic Hard Negative Optimizer, utilizing dual prompts to construct a more challenging optimization task on base classes for enhancement. For interpretability, we prove the feature channel invariance of the prompt vector during the optimization process, providing theoretical support for the Weighting-Decoupling of DPC. Extensive experiments on multiple backbones demonstrate that DPC can significantly improve base performance without introducing any external knowledge beyond the base classes, while maintaining generalization to new classes. Code is available at: https://github.com/JREion/DPC.

Effectively leveraging private datasets remains a significant challenge in developing foundation models. Federated Learning (FL) has recently emerged as a collaborative framework that enables multiple users to fine-tune these models while mitigating data privacy risks. Meanwhile, Low-Rank Adaptation (LoRA) offers a resource-efficient alternative for fine-tuning foundation models by dramatically reducing the number of trainable parameters. This survey examines how LoRA has been integrated into federated fine-tuning for foundation models—an area we term FedLoRA—by focusing on three key challenges: distributed learning, heterogeneity, and efficiency. We further categorize existing work based on the specific methods used to address each challenge. Finally, we discuss open research questions and highlight promising directions for future investigation, outlining the next steps for advancing FedLoRA.

Volumetric videos are essential for immersive applications due to their engaging and realistic experiences. However, streaming them in real time over constrained, fluctuating networks remains challenging. Progressive streaming is an effective method to mitigate this issue by gradually enhancing video quality through incremental data transmission. However, existing progressive volumetric streaming solutions often rely on specific compression algorithms or require codec modifications, leading to poor compatibility with standard codecs. In this paper, we propose P2VS, a progressive partition-based volumetric video streaming framework, to achieve codec-independent progressive streaming. Specifically, P2VS leverages the unique structure of point cloud-based volumetric video to incrementally enhance video quality without being constrained by specific compression algorithms. Moreover, we propose adaptive streaming algorithms under this framework to enhance the quality of experience (QoE). Extensive simulations demonstrate that P2VS improves QoE by 21% on average compared to non-progressive streaming schemes. It also achieves better bandwidth efficiency and full compatibility with standard codecs. A prototype is built to verify the feasibility of P2VS.

Can we build accurate world models out of large language models (LLMs)? How can world models benefit LLM agents? The gap between the prior knowledge of LLMs and the specified environment's dynamics usually bottlenecks LLMs' performance as world models. To bridge the gap, we propose a training-free "world alignment" that learns an environment's symbolic knowledge complementary to LLMs. The symbolic knowledge covers action rules, knowledge graphs, and scene graphs, which are extracted by LLMs from exploration trajectories and encoded into executable codes to regulate LLM agents' policies. We further propose an RL-free, model-based agent "WALL-E" through the model-predictive control (MPC) framework. Unlike classical MPC requiring costly optimization on the fly, we adopt an LLM agent as an efficient look-ahead optimizer of future steps' actions by interacting with the neurosymbolic world model. While the LLM agent's strong heuristics make it an efficient planner in MPC, the quality of its planned actions is also secured by the accurate predictions of the aligned world model. They together considerably improve learning efficiency in a new environment. On open-world challenges in Mars (Minecraft like) and ALFWorld (embodied indoor environments), WALL-E significantly outperforms existing methods, e.g., surpassing baselines in Mars by 16.1%–51.6% of success rate and by at least 61.7% in score. In ALFWorld, it achieves a new record 98% success rate after only 4 iterations.

Real-time video analytics demand intensive computing resources, often exceeding device capabilities. Heterogeneous computing resources like CPU and GPU, usually work collaboratively to ensure real-time performance. GPU manage data-intensive computing, while CPU handle instruction-intensive tasks. However, analytics accuracy can degrade because of the imbalance between CPU-GPU workloads and resources, and static resources in most systems limit adaptability to dynamic workloads. In addition to CPU-GPU resource control, accuracy is highly dependent on video analytics configuration including resolution, frame rate, and model selection. In this paper, we propose Gemini+, a hardware-accelerated video analytics pipeline empowered by dual-image FPGAs. It enables flexible CPU-GPU resource adaptation by providing near-instantaneous switching between two pre-configured images. We investigate CPU-GPU resource control and video analytics configuration adaptation in a dual-image FPGA-based video analytics pipeline. We study and formulate two problems of practical importance, a single-camera and multi-camera dual computing resource control problem, i.e., SC-DCRC and MC-DCRC. We analyze the problem complexity and develop optimal and sub-optimal algorithms. We evaluate our algorithms through simulation and build a prototype for verification. The results show that Gemini+ can improve analytic accuracy by 25% compared to fixed CPU-GPU resource systems and 88% compared to GPU-dominant systems.

While large language models (LLMs) excel in a simulated world of texts, they struggle to interact with the more realistic world without perceptions of other modalities such as visual or audio signals. Although vision-language models (VLMs) integrate LLM modules (1) aligned with static image features, and (2) may possess prior knowledge of world dynamics (as demonstrated in the text world), they have not been trained in an embodied visual world and thus cannot align with its dynamics. On the other hand, training an embodied agent in a noisy visual world without expert guidance is often chal-lenging and inefficient. In this paper, we train a VLM agent living in a visual world using an LLM agent excelling in a parallel text world. Specifically, we distill LLM's reflection outcomes (improved actions by analyzing mistakes) in a text world's tasks to finetune the VLM on the same tasks of the visual world, resulting in an Embodied Multi-Modal Agent (EMMA) quickly adapting to the visual world dy-namics. Such cross-modality imitation learning between the two parallel worlds is achieved by a novel DAgger-DPO algorithm, enabling EMMA to generalize to a broad scope of new tasks without any further guidance from the LLM expert. Extensive evaluations on the ALFWorld benchmark's diverse tasks highlight EMMA's superior performance to SOTA VLM-based agents, e.g., 20%-70% improvement in the success rate.

Yang Li, Canran Xu, Guodong Long, Tao Shen, Chongyang Tao, Jing Jiang. Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

This paper demonstrates that pre-trained language models (PLMs) are strong foundation models for on-device meteorological variable modeling. We present LM-Weather, a generic approach to taming PLMs, that have learned massive sequential knowledge from the universe of natural language databases, to acquire an immediate capability to obtain highly customized models for heterogeneous meteorological data on devices while keeping high efficiency. Concretely, we introduce a lightweight personalized adapter into PLMs and endows it with weather pattern awareness. During communication between clients and the server, low-rank-based transmission is performed to effectively fuse the global knowledge among devices while maintaining high communication efficiency and ensuring privacy. Experiments on real-wold dataset show that LM-Weather outperforms the state-of-the-art results by a large margin across various tasks (e.g., forecasting and imputation at different scales). We provide extensive and in-depth analyses experiments, which verify that LM-Weather can (1) indeed leverage sequential knowledge from natural language to accurately handle meteorological sequence, (2) allows each devices obtain highly customized models under significant heterogeneity, and (3) generalize under data-limited and out-of-distribution (OOD) scenarios.

Recently, foundation models, particularly large language models (LLMs), have demonstrated an impressive ability to adapt to various tasks by fine-tuning diverse instruction data. Notably, federated foundation models (FedFM) emerge as a privacy preservation method to fine-tune models collaboratively under federated learning (FL) settings by leveraging many distributed datasets with non-IID data. To alleviate communication and computation overhead, parameter-efficient methods are introduced for efficiency, and some research adapted personalization methods to FedFM for better user preferences alignment. However, a critical gap in existing research is the neglect of test-time distribution shifts in real-world applications, and conventional methods for test-time distribution shifts in personalized FL are less effective for FedFM due to their failure to adapt to complex distribution shift scenarios and the requirement to train all parameters. To bridge this gap, we refine the setting in FedFM, termed test-time personalization, which aims to learn personalized federated foundation models on clients while effectively handling test-time distribution shifts simultaneously. To address challenges in this setting, we explore a simple yet effective solution, a Federated Dual-Personalizing Adapter (FedDPA) architecture. By co-working with a foundation model, a global adapter and a local adapter jointly tackle the test-time distribution shifts and client-specific personalization. Additionally, we introduce an instance-wise dynamic weighting mechanism that dynamically integrates the global and local adapters for each test instance during inference, facilitating effective test-time personalization. The effectiveness of the proposed method has been evaluated on benchmark datasets across different NLP tasks.

Offline reinforcement learning (RL) harnesses the power of massive datasets for resolving sequential decision problems. Most existing papers only discuss defending against out-of-distribution (OOD) actions while we investigate a broader issue, the false correlations between epistemic uncertainty and decision-making, an essential factor that causes suboptimality. In this paper, we propose falSe COrrelation REduction (SCORE) for offline RL, a practically effective and theoretically provable algorithm. We empirically show that SCORE achieves the SoTA performance with 3.1x acceleration on various tasks in a standard benchmark (D4RL). The proposed algorithm introduces an annealing behavior cloning regularizer to help produce a high-quality estimation of uncertainty which is critical for eliminating false correlations from suboptimality. Theoretically, we justify the rationality of the proposed method and prove its convergence to the optimal policy with a sublinear rate under mild assumptions.

Graph neural networks (GNNs) have shown their superiority in modeling graph data. Owing to the advantages of federated learning, federated graph learning (FGL) enables clients to train strong GNN models in a distributed manner without sharing their private data. A core challenge in federated systems is the non-IID problem, which also widely exists in real-world graph data. For example, local data of clients may come from diverse datasets or even domains, e.g., social networks and molecules, increasing the difficulty for FGL methods to capture commonly shared knowledge and learn a generalized encoder. From real-world graph datasets, we observe that some structural properties are shared by various domains, presenting great potential for sharing structural knowledge in FGL. Inspired by this, we propose FedStar, an FGL framework that extracts and shares the common underlying structure information for inter-graph federated learning tasks. To explicitly extract the structure information rather than encoding them along with the node features, we define structure embeddings and encode them with an independent structure encoder. Then, the structure encoder is shared across clients while the feature-based knowledge is learned in a personalized way, making FedStar capable of capturing more structure-based domain-invariant information and avoiding feature misalignment issues. We perform extensive experiments over both cross-dataset and cross-domain non-IID FGL settings, demonstrating the superiority of FedStar.

Conventional reinforcement learning (RL) needs an environment to collect fresh data, which is impractical when online interactions are costly. Offline RL provides an alternative solution by directly learning from the previously collected dataset. However, it will yield unsatisfactory performance if the quality of the offline datasets is poor. In this paper, we consider an offline-to-online setting where the agent is first learned from the offline dataset and then trained online, and propose a framework called Adaptive Policy Learning for effectively taking advantage of offline and online data. Specifically, we explicitly consider the difference between the online and offline data and apply an adaptive update scheme accordingly, that is, a pessimistic update strategy for the offline dataset and an optimistic/greedy update scheme for the online dataset. Such a simple and effective method provides a way to mix the offline and online RL and achieve the best of both worlds. We further provide two detailed algorithms for implementing the framework through embedding value or policy-based RL algorithms into it. Finally, we conduct extensive experiments on popular continuous control tasks, and results show that our algorithm can learn the expert policy with high sample efficiency even when the quality of offline dataset is poor, e.g., random dataset.

How to train a generalizable meta-policy by continually learning a sequence of tasks? It is a natural human skill yet challenging to achieve by current reinforcement learning: the agent is expected to quickly adapt to new tasks (plasticity) meanwhile retaining the common knowledge from previous tasks (stability). We address it by "Continual Task Allocation via Sparse Prompting (CoTASP)", which learns over-complete dictionaries to produce sparse masks as prompts extracting a sub-network for each task from a meta-policy network. CoTASP trains a policy for each task by optimizing the prompts and the sub-network weights alternatively. The dictionary is then updated to align the optimized prompts with tasks’ embedding, thereby capturing tasks’ semantic correlations. Hence, relevant tasks share more neurons in the meta-policy network due to similar prompts while cross-task interference causing forgetting is effectively restrained. Given a meta-policy and dictionaries trained on previous tasks, new task adaptation reduces to highly efficient sparse prompting and sub-network finetuning. In experiments, CoTASP achieves a promising plasticity-stability trade-off without storing or replaying any past tasks’ experiences. It outperforms existing continual and multi-task RL methods on all seen tasks, forgetting reduction, and generalization to unseen tasks.

Distribution shift (e.g., task or domain shift) in continual learning (CL) usually results in catastrophic forgetting of previously learned knowledge. Although it can be alleviated by repeatedly replaying buffered data, the every-step replay is time-consuming. In this paper, we study which modules in neural networks are more prone to forgetting by investigating their training dynamics during CL. Our proposed metrics show that only a few modules are more task-specific and sensitive to task change, while others can be shared across tasks as common knowledge. Hence, we attribute forgetting mainly to the former and find that finetuning them only on a small buffer at the end of any CL method can bring non-trivial improvement. Due to the small number of finetuned parameters, such ''Forgetting Prioritized Finetuning (FPF)'' is efficient in computation. We further propose a more efficient and simpler method that entirely removes the every-step replay and replaces them by only $k$-times of FPF periodically triggered during CL. Surprisingly, this ''$k$-FPF'' performs comparably to FPF and outperforms the SOTA CL methods but significantly reduces their computational overhead and cost. In experiments on several benchmarks of class- and domain-incremental CL, FPF consistently improves existing CL methods by a large margin, and $k$-FPF further excels in efficiency without degrading the accuracy. We also empirically studied the impact of buffer size, epochs per task, and finetuning modules on the cost and accuracy of our methods.

To tackle the global climate challenge, it urgently needs to develop a collaborative platform for comprehensive weather forecasting on large-scale meteorological data. Despite urgency, heterogeneous meteorological sensors across countries and regions, inevitably causing multivariate heterogeneity and data exposure, become the main barrier. This paper develops a foundation model across regions capable of understanding complex meteorological data and providing weather forecasting. To relieve the data exposure concern across regions, a novel federated learning approach has been proposed to collaboratively learn a brand-new spatio-temporal Transformer-based foundation model across participants with heterogeneous meteorological data. Moreover, a novel prompt learning mechanism has been adopted to satisfy low-resourced sensors' communication and computational constraints. The effectiveness of the proposed method has been demonstrated on classical weather forecasting tasks using three meteorological datasets with multivariate time series.

Heterogeneous federated learning without assuming any structure is challenging due to the conflicts among non-identical data distributions of clients. In practice, clients often comprise near-homogeneous clusters so training a server-side model per cluster mitigates the conflicts. However, FL with client clustering often suffers from “clustering collapse'', i.e., one cluster's model excels on increasing clients, and reduces to single-model FL. Moreover, cluster-wise models hinder knowledge sharing between clusters and each model depends on fewer clients. Furthermore, the static clustering assumption on data may not hold for dynamically changing models, which are sensitive to cluster imbalance/initialization or outliers. To address these challenges, we propose ''Clustered Additive Modeling (CAM)'', which applies a globally shared model $\Theta_g$ on top of the cluster-wise models $\Theta_{1:K}$, i.e., $y=h(x;\Theta_g)+f(x;\Theta_k)$ for clients of cluster-$k$. The global model captures the features shared by all clusters so $\Theta_{1:K}$ are enforced to focus on the difference among clusters. To train CAM, we develop a novel Fed-CAM algorithm that alternates between client clustering and training global/cluster models to predict the residual of each other. We can easily modify any existing clustered FL methods by CAM and significantly improve their performance without ‘’clustering collapse'' in different non-IID settings. We also provide a convergence analysis of Fed-CAM algorithm.

Unsupervised video domain adaptation is a practical yet challenging task. In this work, for the first time, we tackle it from a disentanglement view. Our key idea is to handle the spatial and temporal domain divergence separately through disentanglement. Specifically, we consider the generation of cross-domain videos from two sets of latent factors, one encoding the static information and another encoding the dynamic information. A Transfer Sequential VAE (TranSVAE) framework is then developed to model such generation. To better serve for adaptation, we propose several objectives to constrain the latent factors. With these constraints, the spatial divergence can be readily removed by disentangling the static domain-specific information out, and the temporal divergence is further reduced from both frame- and video-levels through adversarial learning. Extensive experiments on the UCF-HMDB, Jester, and Epic-Kitchens datasets verify the effectiveness and superiority of TranSVAE compared with several state-of-the-art approaches.

Abstract Multivariate time series classification is a critical problem in data mining with broad applications. It requires harnessing the inter-relationship of multiple variables and various ranges of temporal dependencies to assign the correct classification label of the time series. Multivariate time series may come from a wide range of sources and be used in various scenarios, bringing the classifier challenge of temporal representation learning. We propose a novel convolutional neural network architecture called Attentional Gated Res2Net for multivariate time series classification. Our model uses hierarchical residual-like connections to achieve multi-scale receptive fields and capture multi-granular temporal information. The gating mechanism enables the model to consider the relations between the feature maps extracted by receptive fields of multiple sizes for information fusion. Further, we propose two types of attention modules, channel-wise attention and block-wise attention, to better leverage the multi-granular temporal patterns. Our experimental results on 14 benchmark multivariate time-series datasets show that our model outperforms several baselines and state-of-the-art methods by a large margin. Our model outperforms the SOTA by a large margin, the classification accuracy of our model is 10.16% better than the SOTA model. Besides, we demonstrate that our model improves the performance of existing models when used as a plugin. Further, based on our experiments and analysis, we provide practical advice on applying our model to a new problem.

Graph convolutional networks are becoming indispensable for deep learning from graph-structured data. Most of the existing graph convolutional networks share two big shortcomings. First, they are essentially low-pass filters, thus the potentially useful middle and high frequency band of graph signals are ignored. Second, the bandwidth of existing graph convolutional filters is fixed. Parameters of a graph convolutional filter only transform the graph inputs without changing the curvature of a graph convolutional filter function. In reality, we are uncertain about whether we should retain or cut off the frequency at a certain point unless we have expert domain knowledge. In this paper, we propose Automatic Graph Convolutional Networks (AutoGCN) to capture the full spectrum of graph signals and automatically update the bandwidth of graph convolutional filters. While it is based on graph spectral theory, our AutoGCN is also localized in space and has a spatial form. Experimental results show that AutoGCN achieves significant improvement over baseline methods which only work as low-pass filters.

Abstract Personalized decision-making can be implemented in a Federated learning (FL) framework that can collaboratively train a decision model by extracting knowledge across intelligent clients, e.g. smartphones or enterprises. FL can mitigate the data privacy risk of collaborative training since it merely collects local gradients from users without access to their data. However, FL is fragile in the presence of statistical heterogeneity that is commonly encountered in personalized decision making, e.g., non-IID data over different clients. Existing FL approaches usually update a single global model to capture the shared knowledge of all users by aggregating their gradients, regardless of the discrepancy between their data distributions. By comparison, a mixture of multiple global models could capture the heterogeneity across various clients if assigning the client to different global models (i.e., centers) in FL. To this end, we propose a novel multi-center aggregation mechanism to cluster clients using their models’ parameters. It learns multiple global models from data as the cluster centers, and simultaneously derives the optimal matching between users and centers. We then formulate it as an optimization problem that can be efficiently solved by a stochastic expectation maximization (EM) algorithm. Experiments on multiple benchmark datasets of FL show that our method outperforms several popular baseline methods. The experimental source codes are publicly available on the Github repository (GitHub repository: https://github.com/mingxuts/multi-center-fed-learning ).

Heterogeneity across clients in federated learning (FL) usually hinders the optimization convergence and generalization performance when the aggregation of clients' knowledge occurs in the gradient space. For example, clients may differ in terms of data distribution, network latency, input/output space, and/or model architecture, which can easily lead to the misalignment of their local gradients. To improve the tolerance to heterogeneity, we propose a novel federated prototype learning (FedProto) framework in which the clients and server communicate the abstract class prototypes instead of the gradients. FedProto aggregates the local prototypes collected from different clients, and then sends the global prototypes back to all clients to regularize the training of local models. The training on each client aims to minimize the classification error on the local data while keeping the resulting local prototypes sufficiently close to the corresponding global ones. Moreover, we provide a theoretical analysis to the convergence rate of FedProto under non-convex objectives. In experiments, we propose a benchmark setting tailored for heterogeneous FL, with FedProto outperforming several recent FL approaches on multiple datasets.

Multivariate time series classification is a critical problem in data mining with broad applications. We design a novel convolutional neural network architecture, Attentional Gated Res2Net, for robust multivariate time series classification. AGRes2Net uses hierarchical residual-like connections to achieve multi-scale receptive fields and to capture multi- granular temporal patterns. It further employs the gated mechanism to harness inter-relationship between feature maps. We propose two types of attention modules, namely channel-wise attention and block-wise attention, to leverage the multi-granular temporal patterns. Our experiments on six benchmark datasets demonstrate that AGRes2Net not only outperforms several baselines and state-of-the-art methods but also improves the classification accuracy of existing models when used as a plug-in.

Positive and Unlabeled learning (PU learning) trains a binary classifier based on only positive (P) and unlabeled (U) data, where the unlabeled data contains positive or negative samples. Previous importance reweighting approaches treat all unlabeled samples as weighted negative samples, achieving state-of-the-art performance. However, in this paper, we surprisingly find that the classifier could misclassify negative samples in U data as positive ones at the late training stage by weight adjustment. Motivated by this discovery, we leverage Semi-Supervised Learning (SSL) to address this performance degradation problem. To this end, we propose a novel SSL-based framework to tackle PU learning. Firstly, we introduce the dynamic increasing sampling strategy to progressively select both negative and positive samples from U data. Secondly, we adopt MixMatch to take full advantage of the unchosen samples in U data. Finally, we propose the Co-learning strategy that iteratively trains two independent networks with the selected samples to avoid the confirmation bias. Experimental results on four benchmark datasets demonstrate the effectiveness and superiority of our approach when compared with other state-of-the-art methods.

Reinforcement learning (RL) is inefficient on long-horizon tasks due to sparse rewards and its policy can be fragile to slightly perturbed environments. We address these challenges via a curriculum of tasks with coupled environments, generated by two policies trained jointly with RL: (1) a co-operative planning policy recursively decomposing a hard task into a coarse-to-fine sub-task tree; and (2) an adversarial policy modifying the environment in each sub-task. They are complementary to acquire more informative feedback for RL: (1) provides dense reward of easier sub-tasks while (2) modifies sub-tasks’ environments to be more challenging and diverse. Conversely, they are trained by RL’s dense feedback on sub-tasks so their generated curriculum keeps adaptive to RL’s progress. The sub-task tree enables an easy-to-hard curriculum for every policy: its top-down construction gradually increases sub-tasks the planner needs to generate, while the adversarial training between the environment and RL follows a bottom-up traversal that starts from a dense sequence of easier sub-tasks allowing more frequent environment changes. We compare EAT-C with RL/planning targeting similar problems and methods with environment generators or adversarial agents. Extensive experiments on diverse tasks demonstrate the advantages of our method on improving RL’s efficiency and generalization.

Knowledge sharing and model personalization are two key components in the conceptual framework of personalized federated learning (PFL). Existing PFL methods focus on proposing new model personalization mechanisms while simply implementing knowledge sharing by aggregating models from all clients, regardless of their relation graph. This paper aims to enhance the knowledge-sharing process in PFL by leveraging the graph-based structural information among clients. We propose a novel structured federated learning (SFL) framework to learn both the global and personalized models simultaneously using client-wise relation graphs and clients' private data. We cast SFL with graph into a novel optimization problem that can model the client-wise complex relations and graph-based structural topology by a unified framework. Moreover, in addition to using an existing relation graph, SFL could be expanded to learn the hidden relations among clients. Experiments on traffic and image benchmark datasets can demonstrate the effectiveness of the proposed method.

Federated Learning (FL) is a machine learning paradigm that allows decentralized clients to learn collaboratively without sharing their private data. However, excessive computation and communication demands pose challenges to current FL frameworks, especially when training large-scale models. To prevent these issues from hindering the deployment of FL systems, we propose a lightweight framework where clients jointly learn to fuse the representations generated by multiple fixed pre-trained models rather than training a large-scale model from scratch. This leads us to a more practical FL problem by considering how to capture more client-specific and class-relevant information from the pre-trained models and jointly improve each client's ability to exploit those off-the-shelf models. Here, we design a Federated Prototype-wise Contrastive Learning (FedPCL) approach which shares knowledge across clients through their class prototypes and builds client-specific representations in a prototype-wise contrastive manner. Sharing prototypes rather than learnable model parameters allows each client to fuse the representations in a personalized way while keeping the shared knowledge in a compact form for efficient communication. We perform a thorough evaluation of the proposed FedPCL in the lightweight framework, measuring and visualizing its ability to fuse various pre-trained models on popular FL datasets.

We study many-class few-shot (MCFS) problem in both supervised learning and meta-learning settings. Compared to the well-studied many-class many-shot and few-class few-shot problems, the MCFS problem commonly occurs in practical applications but has been rarely studied in previous literature. It brings new challenges of distinguishing between many classes given only a few training samples per class. In this article, we leverage the class hierarchy as a prior knowledge to train a coarse-to-fine classifier that can produce accurate predictions for MCFS problem in both settings. The propose model, “memory-augmented hierarchical-classification network (MahiNet)”, performs coarse-to-fine classification where each coarse class can cover multiple fine classes. Since it is challenging to directly distinguish a variety of fine classes given few-shot data per class, MahiNet starts from learning a classifier over coarse-classes with more training data whose labels are much cheaper to obtain. The coarse classifier reduces the searching range over the fine classes and thus alleviates the challenges from “many classes”. On architecture, MahiNet first deploys a convolutional neural network (CNN) to extract features. It then integrates a memory-augmented attention module and a multi-layer perceptron (MLP) together to produce the probabilities over coarse and fine classes. While the MLP extends the linear classifier, the attention module extends the KNN classifier, both together targeting the “few-shot” problem. We design several training strategies of MahiNet for supervised learning and meta-learning. In addition, we propose two novel benchmark datasets “mcfsImageNet” (as a subset of ImageNet) and “mcfsOmniglot” (re-splitted Omniglot) specially designed for MCFS problem. In experiments, we show that MahiNet outperforms several state-of-the-art models (e.g., prototypical networks and relation networks) on MCFS problems in both supervised learning and meta-learning.

Sequential diagnosis prediction on the Electronic Health Record (EHR) has been proven crucial for predictive analytics in the medical domain. EHR data, sequential records of a patient’s interactions with healthcare systems, has numerous inherent characteristics of temporality, irregularity and data insufficiency. Some recent works train healthcare predictive models by making use of sequential information in EHR data, but they are vulnerable to irregular, temporal EHR data with the states of admission/discharge from hospital, and insufficient data. To mitigate this, we propose an end-to-end robust transformer-based model called SETOR, which exploits neural ordinary differential equation to handle both irregular intervals between a patient’s visits with admitted timestamps and length of stay in each visit, to alleviate the limitation of insufficient data by integrating medical ontology, and to capture the dependencies between the patient’s visits by employing multi-layer transformer blocks. Experiments conducted on two real-world healthcare datasets show that, our sequential diagnoses prediction model SETOR not only achieves better predictive results than previous state-of-the-art approaches, irrespective of sufficient or insufficient training data, but also derives more interpretable embeddings of medical codes. The experimental codes are available at the GitHub repository1.1Github repository: https://github.com/Xueping/SETOR

Goal-conditioned reinforcement learning (RL) usually suffers from sparse reward and inefficient exploration in long-horizon tasks. Planning can find the shortest path to a distant goal that provides dense reward/guidance but is inaccurate without a precise environment model. We show that RL and planning can collaboratively learn from each other to overcome their own drawbacks. In ''CO-PILOT'', a learnable path-planner and an RL agent produce dense feedback to train each other on a curriculum of tree-structured sub-tasks. Firstly, the planner recursively decomposes a long-horizon task to a tree of sub-tasks in a top-down manner, whose layers construct coarse-to-fine sub-task sequences as plans to complete the original task. The planning policy is trained to minimize the RL agent's cost of completing the sequence in each layer from top to bottom layers, which gradually increases the sub-tasks and thus forms an easy-to-hard curriculum for the planner. Next, a bottom-up traversal of the tree trains the RL agent from easier sub-tasks with denser rewards on bottom layers to harder ones on top layers and collects its cost on each sub-task train the planner in the next episode. CO-PILOT repeats this mutual training for multiple episodes before switching to a new task, so the RL agent and planner are fully optimized to facilitate each other's training. We compare CO-PILOT with RL (SAC, HER, PPO), planning (RRT*, NEXT, SGT), and their combination (SoRB) on navigation and continuous control tasks. CO-PILOT significantly improves the success rate and sample efficiency.

Neighborhood aggregation algorithms like spectral graph convolutional networks (GCNs) formulate graph convolutions as a symmetric Laplacian smoothing operation to aggregate the feature information of one node with that of its neighbors. While they have achieved great success in semi-supervised node classification on graphs, current approaches suffer from the over-smoothing problem when the depth of the neural networks increases, which always leads to a noticeable degradation of performance. To solve this problem, we present graph convolutional ladder-shape networks (GCLN), a novel graph neural network architecture that transmits messages from shallow layers to deeper layers to overcome the over-smoothing problem and dramatically extend the scale of the neural networks with improved performance. We have validated the effectiveness of proposed GCLN at a node-wise level with a semi-supervised task (node classification) and an unsupervised task (node clustering), and at a graph-wise level with graph classification by applying a differentiable pooling operation. The proposed GCLN outperforms original GCNs, deep GCNs and other state-of-the-art GCN-based models for all three tasks, which were designed from various perspectives on six real-world benchmark data sets.

Distantly supervised relation extraction intrinsically suffers from noisy labels due to the strong assumption of distant supervision. Most prior works adopt a selective attention mechanism over sentences in a bag to denoise from wrongly labeled data, which however could be incompetent when there is only one sentence in a bag. In this paper, we propose a brand-new light-weight neural framework to address the distantly supervised relation extraction problem and alleviate the defects in previous selective attention framework. Specifically, in the proposed framework, 1) we use an entity-aware word embedding method to integrate both relative position information and head/tail entity embeddings, aiming to highlight the essence of entities for this task; 2) we develop a self-attention mechanism to capture the rich contextual dependencies as a complement for local dependencies captured by piecewise CNN; and 3) instead of using selective attention, we design a pooling-equipped gate, which is based on rich contextual representations, as an aggregator to generate bag-level representation for final relation classification. Compared to selective attention, one major advantage of the proposed gating mechanism is that, it performs stably and promisingly even if only one sentence appears in a bag and thus keeps the consistency across all training examples. The experiments on NYT dataset demonstrate that our approach achieves a new state-of-the-art performance in terms of both AUC and top-n precision metrics.

The goal of zero-shot learning (ZSL) is to train a model to classify samples of classes that were not seen during training. To address this challenging task, most ZSL methods relate unseen test classes to seen(training) classes via a pre-defined set of attributes that can describe all classes in the same semantic space, so the knowledge learned on the training classes can be adapted to unseen classes. In this paper, we aim to optimize the attribute space for ZSL by training a propagation mechanism to refine the semantic attributes of each class based on its neighbors and related classes on a graph of classes. We show that the propagated attributes can produce classifiers for zero-shot classes with significantly improved performance in different ZSL settings. The graph of classes is usually free or very cheap to acquire such as WordNet or ImageNet classes. When the graph is not provided, given pre-defined semantic embeddings of the classes, we can learn a mechanism to generate the graph in an end-to-end manner along with the propagation mechanism. However, this graph-aided technique has not been well-explored in the literature. In this paper, we introduce the “attribute propagation network (APNet)”, which is composed of 1) a graph propagation model generating attribute vector for each class and 2) a parameterized nearest neighbor (NN) classifier categorizing an image to the class with the nearest attribute vector to the image's embedding. For better generalization over unseen classes, different from previous methods, we adopt a meta-learning strategy to train the propagation mechanism and the similarity metric for the NN classifier on multiple sub-graphs, each associated with a classification task over a subset of training classes. In experiments with two zero-shot learning settings and five benchmark datasets, APNet achieves either compelling performance or new state-of-the-art results.

Many graph embedding approaches have been proposed for knowledge graph completion via link prediction.Among those, translating embedding approaches enjoy the advantages of lightweight structure, high efficiency and great interpretability.Especially when extended to complex vector space, they show the capability in handling various relation patterns including symmetry, antisymmetry, inversion and composition.However, previous translating embedding approaches defined in complex vector space suffer from two main issues: 1) representing and modeling capacities of the model are limited by the translation function with rigorous multiplication of two complex numbers; and 2) embedding ambiguity caused by one-to-many relations is not explicitly alleviated.In this paper, we propose a relation-adaptive translation function built upon a novel weighted product in complex space, where the weights are learnable, relation-specific and independent to embedding size.The translation function only requires eight more scalar parameters each relation, but improves expressive power and alleviates embedding ambiguity problem.Based on the function, we then present our Relation-adaptive translating Embedding (RatE) approach to score each graph triple.Moreover, a novel negative sampling method is proposed to utilize both prior knowledge and self-adversarial learning for effective optimization.Experiments verify RatE achieves state-of-the-art performance on four link prediction benchmarks.

Wrong labeling problem and long-tail relations are two main challenges caused by distant supervision in relation extraction.Recent works alleviate the wrong labeling by selective attention via multi-instance learning, but cannot well handle long-tail relations even if hierarchies of the relations are introduced to share knowledge.In this work, we propose a novel neural network, Collaborating Relation-augmented Attention (CoRA), to handle both the wrong labeling and long-tail relations.Particularly, we first propose relation-augmented attention network as base model.It operates on sentence bag with a sentence-to-relation attention to minimize the effect of wrong labeling.Then, facilitated by the proposed base model, we introduce collaborating relation features shared among relations in the hierarchies to promote the relation-augmenting process and balance the training data for long-tail relations.Besides the main training objective to predict the relation of a sentence bag, an auxiliary objective is utilized to guide the relationaugmenting process for a more accurate bag-level representation.In the experiments on the popular benchmark dataset NYT, the proposed CoRA improves the prior state-of-the-art performance by a large margin in terms of Precision@N, AUC and [email protected] analyses verify its superior capability in handling long-tail relations in contrast to the competitors.

Electronic health records (EHRs) are longitudinal records of a patient's interactions with healthcare systems. A patient's EHR data is organized as a three-level hierarchy from top to bottom: patient journey - all the experiences of diagnoses and treatments over a period of time; individual visit - a set of medical codes in a particular visit; and medical code - a specific record in the form of medical codes. As EHRs begin to amass in millions, the potential benefits, which these data might hold for medical research and medical outcome prediction, are staggering - including, for example, predicting future admissions to hospitals, diagnosing illnesses or determining the efficacy of medical treatments. Each of these analytics tasks requires a domain knowledge extraction method to transform the hierarchical patient journey into a vector representation for further prediction procedure. The representations should embed a sequence of visits and a set of medical codes with a specific timestamp, which are crucial to any downstream prediction tasks. Hence, expressively powerful representations are appealing to boost learning performance. To this end, we propose a novel self-attention mechanism that captures the contextual dependency and temporal relationships within a patient's healthcare journey. An end-to-end bidirectional temporal encoder network (BiteNet) then learns representations of the patient's journeys, based solely on the proposed attention mechanism. We have evaluated the effectiveness of our methods on two supervised prediction and two unsupervised clustering tasks with a real-world EHR dataset. The empirical results demonstrate the proposed BiteNet model produces higher-quality representations than state-of-the-art baseline methods.

Graph embedding has shown its effectiveness to represent graph information and capture deep relationships in graph data. Most recent graph embedding methods focus on attributed graphs, since they preserve both structure and content information in the network. However, corruption can exist in the graph structure as well as the node content of the graph, and both can lead to inferior embedding results. Unfortunately, few existing graph embedding algorithms have considered the corruption problem, and to the best of our knowledge, none has studied structural corruption in attributed graphs, including missing and redundant edges. This field is difficult for previous methods, mainly due to two challenges: (1) the existence of various corruption causes has made it difficult to recognize corruptions in graphs, and (2) the complexity of graph-structured data has increased the difficulty of handling corruption therein for graph embedding methods. These facts lead us here to propose a novel autoencoder-based graph embedding approach, which is robust against structural corruption. Our idea comes from the recent discovery of memorization effects in deep learning. Namely, deep neural networks prefer to fit clean data first, before they over-fit corrupted data. Specifically, we train two autoencoders simultaneously and let them learn the reliability of the edges in the graph from each other. The two autoencoders would evaluate the edges according to their reconstructed structure and manipulate this by devaluing those distrusted edges to update the structure information. The updated structure would be used further in the next iteration as the ground-truth of its peer-network. Experiments on different versions of real-world graphs show state-of-the-art results and demonstrate the robustness of our model against structural corruption.

Many algorithms for Knowledge-Based Question Answering (KBQA) depend on semantic parsing, which translates a question to its logical form. When only weak supervision is provided, it is usually necessary to search valid logical forms for model training. However, a complex question typically involves a huge search space, which creates two main problems: 1) the solutions limited by computation time and memory usually reduce the success rate of the search, and 2) spurious logical forms in the search results degrade the quality of training data. These two problems lead to a poorly-trained semantic parsing model. In this work, we propose an effective search method for weakly supervised KBQA based on operator prediction for questions. With search space constrained by predicted operators, sufficient search paths can be explored, more valid logical forms can be derived, and operators possibly causing spurious logical forms can be avoided. As a result, a larger proportion of questions in a weakly supervised training set are equipped with logical forms, and fewer spurious logical forms are generated. Such high-quality training data directly contributes to a better semantic parsing model. Experimental results on one of the largest KBQA datasets (i.e., CSQA) verify the effectiveness of our approach and deliver a new state-of-the-art performance.

Modeling multivariate time series has long been a subject that has attracted researchers from a diverse range of fields including economics, finance, and traffic. A basic assumption behind multivariate time series forecasting is that its variables depend on one another but, upon looking closely, it is fair to say that existing methods fail to fully exploit latent spatial dependencies between pairs of variables. In recent years, meanwhile, graph neural networks (GNNs) have shown high capability in handling relational dependencies. GNNs require well-defined graph structures for information propagation which means they cannot be applied directly for multivariate time series where the dependencies are not known in advance. In this paper, we propose a general graph neural network framework designed specifically for multivariate time series data. Our approach automatically extracts the uni-directed relations among variables through a graph learning module, into which external knowledge like variable attributes can be easily integrated. A novel mix-hop propagation layer and a dilated inception layer are further proposed to capture the spatial and temporal dependencies within the time series. The graph learning, graph convolution, and temporal convolution modules are jointly learned in an end-to-end framework. Experimental results show that our proposed model outperforms the state-of-the-art baseline methods on 3 of 4 benchmark datasets and achieves on-par performance with other approaches on two traffic datasets which provide extra structural information.

Imbalanced learning (IL), i.e., learning unbiased models from class-imbalanced data, is a challenging problem. Typical IL methods including resampling and reweighting were designed based on some heuristic assumptions. They often suffer from unstable performance, poor applicability, and high computational cost in complex tasks where their assumptions do not hold. In this paper, we introduce a novel ensemble IL framework named MESA. It adaptively resamples the training set in iterations to get multiple classifiers and forms a cascade ensemble model. MESA directly learns the sampling strategy from data to optimize the final metric beyond following random heuristics. Moreover, unlike prevailing meta-learning-based IL solutions, we decouple the model-training and meta-training in MESA by independently train the meta-sampler over task-agnostic meta-data. This makes MESA generally applicable to most of the existing learning models and the meta-sampler can be efficiently applied to new tasks. Extensive experiments on both synthetic and real-world tasks demonstrate the effectiveness, robustness, and transferability of MESA. Our code is available at https://github.com/ZhiningLiu1998/mesa.

Numerous deep reinforcement learning agents have been proposed, and each of them has its strengths and flaws. In this work, we present a Cooperative Heterogeneous Deep Reinforcement Learning (CHDRL) framework that can learn a policy by integrating the advantages of heterogeneous agents. Specifically, we propose a cooperative learning framework that classifies heterogeneous agents into two classes: global agents and local agents. Global agents are off-policy agents that can utilize experiences from the other agents. Local agents are either on-policy agents or population-based evolutionary algorithms (EAs) agents that can explore the local area effectively. We employ global agents, which are sample-efficient, to guide the learning of local agents so that local agents can benefit from the sample-efficient agents and simultaneously maintain their advantages, e.g., stability. Global agents also benefit from effective local searches. Experimental studies on a range of continuous control tasks from the Mujoco benchmark show that CHDRL achieves better performance compared with state-of-the-art baselines.

Graph embedding aims to transfer a graph into vectors to facilitate subsequent graph-analytics tasks like link prediction and graph clustering. Most approaches on graph embedding focus on preserving the graph structure or minimizing the reconstruction errors for graph data. They have mostly overlooked the embedding distribution of the latent codes, which unfortunately may lead to inferior representation in many cases. In this article, we present a novel adversarially regularized framework for graph embedding. By employing the graph convolutional network as an encoder, our framework embeds the topological information and node content into a vector representation, from which a graph decoder is further built to reconstruct the input graph. The adversarial training principle is applied to enforce our latent codes to match a prior Gaussian or uniform distribution. Based on this framework, we derive two variants of the adversarial models, the adversarially regularized graph autoencoder (ARGA) and its variational version, and adversarially regularized variational graph autoencoder (ARVGA), to learn the graph embedding effectively. We also exploit other potential variations of ARGA and ARVGA to get a deeper understanding of our designs. Experimental results that compared 12 algorithms for link prediction and 20 algorithms for graph clustering validate our solutions.

In longitudinal electronic health records (EHRs), the event records of a patient are distributed over a long period of time and the temporal relations between the events reflect sufficient domain knowledge to benefit prediction tasks such as the rate of inpatient mortality. Medical concept embedding as a feature extraction method that transforms a set of medical concepts with a specific time stamp into a vector, which will be fed into a supervised learning algorithm. The quality of the embedding significantly determines the learning performance over the medical data. In this paper, we propose a medical concept embedding method based on applying a self-attention mechanism to represent each medical concept. We propose a novel attention mechanism which captures the contextual information and temporal relationships between medical concepts. A light-weight neural net, "Temporal Self-Attention Network (TeSAN)", is then proposed to learn medical concept embedding based solely on the proposed attention mechanism. To test the effectiveness of our proposed methods, we have conducted clustering and prediction tasks on two public EHRs datasets comparing TeSAN against five state-of-the-art embedding methods. The experimental results demonstrate that the proposed TeSAN model is superior to all the compared methods. To the best of our knowledge, this work is the first to exploit temporal self-attentive relations between medical events.

A variety of machine learning applications expect to achieve rapid learning from a limited number of labeled data. However, the success of most current models is the result of heavy training on big data. Meta-learning addresses this problem by extracting common knowledge across different tasks that can be quickly adapted to new tasks. However, they do not fully explore weakly-supervised information, which is usually free or cheap to collect. In this paper, we show that weakly-labeled data can significantly improve the performance of meta-learning on few-shot classification. We propose prototype propagation network (PPN) trained on few-shot tasks together with data annotated by coarse-label. Given a category graph of the targeted fine-classes and some weakly-labeled coarse-classes, PPN learns an attention mechanism which propagates the prototype of one class to another on the graph, so that the K-nearest neighbor (KNN) classifier defined on the propagated prototypes results in high accuracy across different few-shot tasks. The training tasks are generated by subgraph sampling, and the training objective is obtained by accumulating the level-wise classification loss on the subgraph. On two benchmarks, PPN significantly outperforms most recent few-shot learning methods in different settings, even when they are also allowed to train on weakly-labeled data.

Graph clustering is a fundamental task which discovers communities or groups in networks. Recent studies have mostly focused on developing deep learning approaches to learn a compact graph embedding, upon which classic clustering methods like k-means or spectral clustering algorithms are applied. These two-step frameworks are difficult to manipulate and usually lead to suboptimal performance, mainly because the graph embedding is not goal-directed, i.e., designed for the specific clustering task. In this paper, we propose a goal-directed deep learning approach, Deep Attentional Embedded Graph Clustering (DAEGC for short). Our method focuses on attributed graphs to sufficiently explore the two sides of information in graphs. By employing an attention network to capture the importance of the neighboring nodes to a target node, our DAEGC algorithm encodes the topological structure and node content in a graph to a compact representation, on which an inner product decoder is trained to reconstruct the graph structure. Furthermore, soft labels from the graph embedding itself are generated to supervise a self-training graph clustering process, which iteratively refines the clustering results. The self-training process is jointly learned and optimized with the graph embedding in a unified framework, to mutually benefit both components. Experimental results compared with state-of-the-art algorithms demonstrate the superiority of our method.

Spatial-temporal graph modeling is an important task to analyze the spatial relations and temporal trends of components in a system. Existing approaches mostly capture the spatial dependency on a fixed graph structure, assuming that the underlying relation between entities is pre-determined. However, the explicit graph structure (relation) does not necessarily reflect the true dependency and genuine relation may be missing due to the incomplete connections in the data. Furthermore, existing methods are ineffective to capture the temporal trends as the RNNs or CNNs employed in these methods cannot capture long-range temporal sequences. To overcome these limitations, we propose in this paper a novel graph neural network architecture, {Graph WaveNet}, for spatial-temporal graph modeling. By developing a novel adaptive dependency matrix and learn it through node embedding, our model can precisely capture the hidden spatial dependency in the data. With a stacked dilated 1D convolution component whose receptive field grows exponentially as the number of layers increases, Graph WaveNet is able to handle very long sequences. These two components are integrated seamlessly in a unified framework and the whole framework is learned in an end-to-end manner. Experimental results on two public traffic network datasets, METR-LA and PEMS-BAY, demonstrate the superior performance of our algorithm.

Graph convolutional networks (GCNs) have recently become one of the most powerful tools for graph analytics tasks in numerous applications, ranging from social networks and natural language processing to bioinformatics and chemoinformatics, thanks to their ability to capture the complex relationships between concepts. At present, the vast majority of GCNs use a neighborhood aggregation framework to learn a continuous and compact vector, then performing a pooling operation to generalize graph embedding for the classification task. These approaches have two disadvantages in the graph classification task: (1)when only the largest sub-graph structure (k-hop neighbor) is used for neighborhood aggregation, a large amount of early-stage information is lost during the graph convolution step; (2) simple average/sum pooling or max pooling utilized, which loses the characteristics of each node and the topology between nodes. In this paper, we propose a novel framework called, dual attention graph convolutional networks (DAGCN) to address these problems. DAGCN automatically learns the importance of neighbors at different hops using a novel attention graph convolution layer, and then employs a second attention component, a self-attention pooling layer, to generalize the graph representation from the various aspects of a matrix graph embedding. The dual attention network is trained in an end-to-end manner for the graph classification task. We compare our model with state-of-the-art graph kernels and other deep learning methods. The experimental results show that our framework not only outperforms other baselines but also achieves a better rate of convergence.

Mobile keyboard suggestion is typically regarded as a word-level language modeling problem. Centralized machine learning techniques require the collection of massive user data for training purposes, which may raise privacy concerns in relation to users' sensitive data. Federated learning (FL) provides a promising approach to learning private language modeling for intelligent personalized keyboard suggestions by training models on distributed clients rather than training them on a central server. To obtain a global model for prediction, existing FL algorithms simply average the client models and ignore the importance of each client during model aggregation. Furthermore, there is no optimization for learning a well-generalized global model on the central server. To solve these problems, we propose a novel model aggregation with an attention mechanism considering the contribution of client models to the global model, together with an optimization technique during server aggregation. Our proposed attentive aggregation method minimizes the weighted distance between the server model and client models by iteratively updating parameters while attending to the distance between the server model and client models. Experiments on two popular language modeling datasets and a social media dataset show that our proposed method outperforms its counterparts in terms of perplexity and communication cost in most settings of comparison.

Electronic health records contain a wealth of information on a patients healthcare over many visits, such as diagnoses, treatments, drugs administered, and so on. The untapped potential of these data in healthcare analytics is vast. However, given that much of medical information is a cause and effect science, new embedding methods are required to ensure the learning representations reflect the comprehensive interplays between medical concepts and their relationships over time. Unlike one-hot encoding, a distributed representation should preserve these complex interactions as high-quality inputs for machine learning-based healthcare analytics tasks. Therefore, we propose a novel attentive dual embedding method called MC2Vec. MC2Vec captures the proximity relationships between medical concepts through a two-step optimization framework that recursively refines the embedding for superior output. The framework comprises a Skip-gram model to generate the initial embedding and an attentive CBOW model to fine-tune the embedding with temporal information gleaned from sequences of patient visits. Experiments with two public datasets demonstrate that MC2Vecs produces embeddings of higher quality than five state-of-the-art methods.

Graphs are ubiquitous for describing and modeling complicated data structures, and graph embedding is an effective solution to learn a mapping from a graph to a low-dimensional vector space while preserving relevant graph characteristics. Most existing graph embedding approaches either embed the topological information and node features separately or learn one regularized embedding with both sources of information, however, they mostly overlook the interdependency between structural characteristics and node features when processing the graph data into the models. Moreover, existing methods only reconstruct the structural characteristics, which are unable to fully leverage the interaction between the topology and the features associated with its nodes during the encoding-decoding procedure. To address the problem, we propose a framework using autoencoder for graph embedding (GED) and its variational version (VEGD). The contribution of our work is two-fold: 1) the proposed frameworks exploit a feature-dependent graph matrix (FGM) to naturally merge the structural characteristics and node features according to their interdependency; and 2) the Graph Convolutional Network (GCN) decoder of the proposed framework reconstructs both structural characteristics and node features, which naturally possesses the interaction between these two sources of information while learning the embedding. We conducted the experiments on three real-world graph datasets such as Cora, Citeseer and PubMed to evaluate our framework and algorithms, and the results outperform baseline methods on both link prediction and graph clustering tasks.

Meta-learning extracts the common knowledge from learning different tasks and uses it for unseen tasks. It can significantly improve tasks that suffer from insufficient training data, e.g., few-shot learning. In most meta-learning methods, tasks are implicitly related by sharing parameters or optimizer. In this paper, we show that a meta-learner that explicitly relates tasks on a graph describing the relations of their output dimensions (e.g., classes) can significantly improve few-shot learning. The graph’s structure is usually free or cheap to obtain but has rarely been explored in previous works. We develop a novel meta-learner of this type for prototype based classification, in which a prototype is generated for each class, such that the nearest neighbor search among the prototypes produces an accurate classification. The meta-learner, called “Gated Propagation Network (GPN)”, learns to propagate messages between prototypes of different classes on the graph, so that learning the prototype of each class benefits from the data of other related classes. In GPN, an attention mechanism aggregates messages from neighboring classes of each class, with a gate choosing between the aggregated message and the message from the class itself. We train GPN on a sequence of tasks from many-shot to few-shot generated by subgraph sampling. During training, it is able to reuse and update previously achieved prototypes from the memory in a life-long learning cycle. In experiments, under different training-test discrepancy and test task generation settings, GPN outperforms recent meta-learning methods on two benchmark datasets. Code of GPN is publicly available at: https://github.com/liulu112601/Gated-Propagation-Net.

Recurrent neural nets (RNN) and convolutional neural nets (CNN) are widely used on NLP tasks to capture the long-term and local dependencies, respectively. Attention mechanisms have recently attracted enormous interest due to their highly parallelizable computation, significantly less training time, and flexibility in modeling dependencies. We propose a novel attention mechanism in which the attention between elements from input sequence(s) is directional and multi-dimensional (i.e., feature-wise). A light-weight neural net, "Directional Self-Attention Network (DiSAN)," is then proposed to learn sentence embedding, based solely on the proposed attention without any RNN/CNN structure. DiSAN is only composed of a directional self-attention with temporal order encoded, followed by a multi-dimensional attention that compresses the sequence into a vector representation. Despite its simple form, DiSAN outperforms complicated RNN models on both prediction quality and time efficiency. It achieves the best test accuracy among all sentence encoding methods and improves the most recent best result by 1.02% on the Stanford Natural Language Inference (SNLI) dataset, and shows state-of-the-art test accuracy on the Stanford Sentiment Treebank (SST), Multi-Genre natural language inference (MultiNLI), Sentences Involving Compositional Knowledge (SICK), Customer Review, MPQA, TREC question-type classification and Subjectivity (SUBJ) datasets.

Graph embedding is an effective method to represent graph data in a low dimensional space for graph analytics. Most existing embedding algorithms typically focus on preserving the topological structure or minimizing the reconstruction errors of graph data, but they have mostly ignored the data distribution of the latent codes from the graphs, which often results in inferior embedding in real-world graph data. In this paper, we propose a novel adversarial graph embedding framework for graph data. The framework encodes the topological structure and node content in a graph to a compact representation, on which a decoder is trained to reconstruct the graph structure. Furthermore, the latent representation is enforced to match a prior distribution via an adversarial training scheme. To learn a robust embedding, two variants of adversarial approaches, adversarially regularized graph autoencoder (ARGA) and adversarially regularized variational graph autoencoder (ARVGA), are developed. Experimental studies on real-world graphs validate our design and demonstrate that our algorithms outperform baselines by a wide margin in link prediction, graph clustering, and graph visualization tasks.

Many natural language processing tasks solely rely on sparse dependencies between a few tokens in a sentence. Soft attention mechanisms show promising performance in modeling local/global dependencies by soft probabilities between every two tokens, but they are not effective and efficient when applied to long sentences. By contrast, hard attention mechanisms directly select a subset of tokens but are difficult and inefficient to train due to their combinatorial nature. In this paper, we integrate both soft and hard attention into one context fusion model, "reinforced self-attention (ReSA)", for the mutual benefit of each other. In ReSA, a hard attention trims a sequence for a soft self-attention to process, while the soft attention feeds reward signals back to facilitate the training of the hard one. For this purpose, we develop a novel hard attention called "reinforced sequence sampling (RSS)", selecting tokens in parallel and trained via policy gradient. Using two RSS modules, ReSA efficiently extracts the sparse dependencies between each pair of selected tokens. We finally propose an RNN/CNN-free sentence-encoding model, "reinforced self-attention network (ReSAN)", solely based on ReSA. It achieves state-of-the-art performance on both the Stanford Natural Language Inference (SNLI) and the Sentences Involving Compositional Knowledge (SICK) datasets.

Over recent decades, globalization has resulted in a steady increase in cross-border financial flows around the world. To build an abstract representation of a real-world financial market situation, we structure the fundamental influences among homogeneous and heterogeneous markets with three types of correlations: the inner-domain correlation between homogeneous markets in various countries, the cross-domain correlation between heterogeneous markets, and the time-series correlation between current and past markets. Such types of correlations in global finance challenge traditional machine learning approaches due to model complexity and nonlinearity. In this paper, we propose a novel cross-domain deep learning approach (Cd-DLA) to learn real-world complex correlations for multiple financial market prediction. Based on recurrent neural networks, which capture the time-series interactions in financial data, our model utilizes the attention mechanism to analyze the inner-domain and cross-domain correlations, and then aggregates all of them for financial forecasting. Experiment results on ten-year financial data on currency and stock markets from three countries prove the performance of our approach over other baselines.

Analyzing accumulated data has recently attracted huge attention for its ability to generate values by identifying useful information and providing an edge in global business competition. However, heterogeneous data and imbalanced class distribution present two major challenges to machine learning with real-world business data. Traditional machine learning algorithms can typically only be applied to standard data sets, which are normally homogeneous and balanced. These algorithms narrow complex data into a homogeneous, a balanced data space an inefficient process that requires a significant amount of pre-processing. In this paper, we focus on an efficient solution to the challenges with heterogeneous and imbalanced data sets that does not require pre-processing. Our approach comprises a novel, unified, end-to-end cost-sensitive hybrid neural network that learns real-world heterogeneous data via a parallel network architecture. A specifically-designed cost-sensitive matrix then automatically generates a robust model for learning minority classifications. And the parameters of both the cost-sensitive matrix and the hybrid neural network are alternately but jointly optimized during training. The results of comparative experiments on six real-world data sets reflecting actual business cases, including insurance fraud detection and mobile customer demographics, indicate that the proposed approach demonstrates superior performance over baseline procedures.

Numerous network representation-based algorithms for network classification have emerged in recent years, but many suffer from two limitations. First, they separate the network representation learning and node classification in networks into two steps, which may result in sub-optimal results because the node representation may not fit the classification model well, and vice versa. Second, they are mostly shallow methods that can only capture the linear and simple relationships in the data. In this paper, we propose an effective deep learning model, Graph Ladder Networks (GLN), for node classification in networks. Our model learns a ladder network which unifies the representation learning and network classification into one single framework by exploiting both labeled and unlabeled nodes in a network. To integrate both structure and node content information in the networks, the most recently developed graph convolution network, is further employed. The experiments on the most popular academic network dataset, Citeseer, demonstrate that our approach reaches outstanding performance compared to other state-of-the-art algorithms.

Graph clustering aims to discovercommunity structures in networks, the task being fundamentally challenging mainly because the topology structure and the content of the graphs are difficult to represent for clustering analysis. Recently, graph clustering has moved from traditional shallow methods to deep learning approaches, thanks to the unique feature representation learning capability of deep learning. However, existing deep approaches for graph clustering can only exploit the structure information, while ignoring the content information associated with the nodes in a graph. In this paper, we propose a novel marginalized graph autoencoder (MGAE) algorithm for graph clustering. The key innovation of MGAE is that it advances the autoencoder to the graph domain, so graph representation learning can be carried out not only in a purely unsupervised setting by leveraging structure and content information, it can also be stacked in a deep fashion to learn effective representation. From a technical viewpoint, we propose a marginalized graph convolutional network to corrupt network node content, allowing node content to interact with network features, and marginalizes the corrupted features in a graph autoencoder context to learn graph feature representations. The learned features are fed into the spectral clustering algorithm for graph clustering. Experimental results on benchmark datasets demonstrate the superior performance of MGAE, compared to numerous baselines.

Most improvements for Naive Bayes (NB) have a common yet important flaw - these algorithms split the modeling of the classifier into two separate stages - the stage of preprocessing (e.g., feature selection and data expansion) and the stage of building the NB classifier. The first stage does not take the NB's objective function into consideration, so the performance of the classification cannot be guaranteed. Motivated by these facts and aiming to improve NB with accurate classification, we present a new learning algorithm called Evolutionary Local Instance Weighted Naive Bayes or ELWNB, to extend NB for classification. ELWNB combines local NB, instance weighted dataset extension and evolutionary algorithms seamlessly. Experiments on 20 UCI benchmark datasets demonstrate that ELWNB significantly outperforms NB and several other improved NB algorithms.

An enterprise social network (ESN) involves diversified user groups from producers, suppliers, logistics, to end consumers, and users have different scales, broad interests, and various objectives, such as advertising, branding, customer relationship management etc. In addition, such a highly diversified network is also featured with rich content, including recruiting messages, advertisements, news release, customer complains etc. Due to such complex nature, an immediate need is to properly organize a chaotic enterprise social network as functional groups, where each group corresponds to a set of peers with business interactions and common objectives, and further understand the business role of each group, such as their common interests and key features differing from other groups. In this paper, we argue that due to unique characteristics of enterprise social networks, simple clustering for ESN nodes or using existing topic discovery methods cannot effectively discover functional groups and understand their roles. Alternatively, we propose CENFLD, which carries out co-clustering on enterprise social networks for functional group discovery and understanding. CENFLD is a co-factorization based framework which combines network topology structures and rich content information, including interactions between nodes and correlations between node content, to discover functional user groups. Because the number of functional groups is highly data driven and hard to estimate, CENFLD employs a hold-out test principle to find the group number optimally complying with the underlying data. Experiments and comparisons, with state-of-the-art approaches, on 13 real-world enterprise/organizational networks validate the performance of CENFLD.

In the on-demand cloud environment, web application providers have the potential to scale virtual resources up or down to achieve cost-effective outcomes. True elasticity and cost-effectiveness in the pay-per-use cloud business model, however, have not yet been achieved. To address this challenge, we propose a novel cloud resource auto-scaling scheme at the virtual machine (VM) level for web application providers. The scheme automatically predicts the number of web requests and discovers an optimal cloud resource demand with cost-latency trade-off. Based on this demand, the scheme makes a resource scaling decision that is up or down or NOP (no operation) in each time-unit re-allocation. We have implemented the scheme on the Amazon cloud platform and evaluated it using three real-world web log datasets. Our experiment results demonstrate that the proposed scheme achieves resource auto-scaling with an optimal cost-latency trade-off, as well as low SLA violations.

Cloud computing has emerging as an extremely popular and cost-effective computational service model using pay-as-you-go executing environments that scale transparently to the user. However, cloud providers should tackle the challenge of configuring their systems to provide maximal performance while minimizing customer's cost of computing resources, which satisfy the customers' various workload requirements. To solve the above challenge, in this paper, we propose an innovative architecture of self-adaptive configuration optimization system which supports dynamic reconfiguration when workloads change. In addition, we develop an optimization algorithm by using genetic algorithm for this system. By using queuing theory and statistic techniques, we model and compute the SLAs metrics which are defined as the fitness function in the optimization algorithm. This optimization system can guide cloud customers to purchase appropriate resources and make decision of deployment configuration such as scale, scheduling and capacity.

Collaborative filtering (CF) techniques have achieved widespread success in E-commerce nowadays. The tremendous growth of the number of customers and products in recent years poses some key challenges for recommender systems in which high quality recommendations are required and more recommendations per second for millions of customers and products need to be performed. Thus, the improvement of scalability and efficiency of collaborative filtering (CF) algorithms become increasingly important and difficult. In this paper, we developed and implemented a scaling-up item-based collaborative filtering algorithm on MapReduce, by splitting the three most costly computations in the proposed algorithm into four Map-Reduce phases, each of which can be independently executed on different nodes in parallel. We also proposed efficient partition strategies not only to enable the parallel computation in each Map-Reduce phase but also to maximize data locality to minimize the communication cost. Experimental results effectively showed the good performance in scalability and efficiency of the item-based CF algorithm on a Hadoop cluster.