Positron emission tomography (PET) is a cornerstone of modern oncologic and neurologic imaging, distinguished by its unique ability to illuminate dynamic metabolic processes that transcend the anatomical focus of traditional imaging technologies. Radiology reports are essential for clinical decision making, yet their manual creation is labor-intensive and time-consuming. Recent advancements of vision-language models (VLMs) have shown strong potential in medical applications, presenting a promising avenue for automating report generation. However, existing applications of VLMs in the medical domain have predominantly focused on structural imaging modalities, while the unique characteristics of molecular PET imaging have largely been overlooked. To bridge the gap, we introduce PET2Rep, a large-scale comprehensive benchmark for evaluation of general and medical VLMs for radiology report generation for PET images. PET2Rep stands out as the first dedicated dataset for PET report generation with metabolic information, uniquely capturing whole-body image-report pairs that cover dozens of organs to fill the critical gap in existing benchmarks and mirror real-world clinical comprehensiveness. In addition to widely recognized natural language generation metrics, we introduce a series of clinical efficiency metrics to evaluate the quality of radiotracer uptake pattern description in key organs in generated reports. We conduct a head-to-head comparison of 30 cutting-edge general-purpose and medical-specialized VLMs. The results show that the current state-of-the-art VLMs perform poorly on PET report generation task, falling considerably short of fulfilling practical needs. Moreover, we identify several key insufficiency that need to be addressed to advance the development in medical applications. We believe PET2Rep will serve as a platform for the development and application of VLMs for PET imaging, accelerating the development of trustworthy reporting tools that can genuinely alleviate radiologist burden and enhance patient care.

Enabling neural networks to learn complex logical constraints and fulfill symbolic reasoning is a critical challenge. Bridging this gap often requires guiding the neural network’s output distribution to move closer to the symbolic constraints. While diffusion models have shown remarkable generative capability across various domains, we employ the powerful architecture to perform neuro-symbolic learning and solve logical puzzles. Our diffusion-based pipeline adopts a two-stage training strategy: the first stage focuses on cultivating basic reasoning abilities, while the second emphasizes systematic learning of logical constraints. To impose hard constraints on neural outputs in the second stage, we formulate the diffusion reasoner as a Markov decision process and innovatively fine-tune it with an improved proximal policy optimization algorithm. We utilize a rule-based reward signal derived from the logical consistency of neural outputs and adopt a flexible strategy to optimize the diffusion reasoner's policy. We evaluate our methodology on some classical symbolic reasoning benchmarks, including Sudoku, Maze, pathfinding and preference learning. Experimental results demonstrate that our approach achieves outstanding accuracy and logical consistency among neural networks.

Protein structure prediction is pivotal for understanding the structure-function relationship of proteins, advancing biological research, and facilitating pharmaceutical development and experimental design. While deep learning methods and the expanded availability of experimental 3D protein structures have accelerated structure prediction, the dynamic nature of protein structures has received limited attention. This study introduces an innovative 4D diffusion model incorporating molecular dynamics (MD) simulation data to learn dynamic protein structures. Our approach is distinguished by the following components: (1) a unified diffusion model capable of generating dynamic protein structures, including both the backbone and side chains, utilizing atomic grouping and side-chain dihedral angle predictions; (2) a reference network that enhances structural consistency by integrating the latent embeddings of the initial 3D protein structures; and (3) a motion alignment module aimed at improving temporal structural coherence across multiple time steps. To our knowledge, this is the first diffusion-based model aimed at predicting protein trajectories across multiple time steps simultaneously. Validation on benchmark datasets demonstrates that our model exhibits high accuracy in predicting dynamic 3D structures of proteins containing up to 256 amino acids over 32 time steps, effectively capturing both local flexibility in stable states and significant conformational changes.

Siming Huang, Tianhao Cheng, Jason Klein Liu, Weidi Xu, Jiaran Hao, Liuyihan Song, Yang Xu, Jian Yang, Jiaheng Liu, Chenchen Zhang, Linzheng Chai, Ruifeng Yuan, Xianzhen Luo, Qiufeng Wang, YuanTao Fan, Qingfu Zhu, Zhaoxiang Zhang, Yang Gao, Jie Fu, Qian Liu, Houyi Li, Ge Zhang, Yuan Qi, Xu Yinghui, Wei Chu, Zili Wang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

With the growing deployment of pre-trained models like Transformers on cloud platforms, privacy concerns about model parameters and inference data are intensifying. Existing Privacy-Preserving Transformer Inference (PPTI) frameworks face the “impossible trinity” of balancing privacy, efficiency, and performance: Secure Multi-Party Computation (SMPC)-based approaches ensure strong privacy but suffer from high computational overhead and performance losses; Conversely, permutation-based methods achieve near-plaintext efficiency and accuracy but compromise privacy by exposing sensitive model parameters and intermediate results. Bridging this gap with a single approach presents substantial challenges, motivating the introduction of CENTAUR, a groundbreaking PPTI framework that seamlessly integrates random permutations and SMPC to address the “impossible trinity”. By designing efficient PPTI algorithms tailored to the structural properties of Transformer models, CENTAUR achieves an unprecedented balance among privacy, efficiency, and performance. Our experiments demonstrate CENTAUR’s ability to resist diverse data reconstruction attacks, achieve plaintext-level inference accuracy, and boost inference speed by 5.0~30.4 times, unlocking new possibilities for secure and efficient AI deployment.

Obstructive sleep apnea-hypopnea syndrome (OS-AHS) is a common sleep disorder caused by upper airway blockage, leading to oxygen deprivation and disrupted sleep. Traditional diagnosis using polysomnography (PSG) is expensive, time-consuming, and uncomfortable. Existing deep learning methods using facial image analysis lack accuracy due to poor facial feature capture and limited sample sizes. To address this, we propose a multimodal dual encoder model that integrates visual and language inputs for automated OSAHS diagnosis. The model balances data using randomOverSampler (ROS), extracts key facial features with attention grids, and converts basic physiological data into meaningful text. Cross attention combines image and text data for better feature extraction, and ordered regression loss ensures stable learning. Our approach improves diagnostic efficiency and accuracy, achieving 91.3% top-1 accuracy in a four class severity classification task, demonstrating state of the art performance. Code is available at https://github.com/luboyan6/VTA-OSAHS.

Positron Emission Tomography (PET) is a powerful molecular imaging tool that plays a crucial role in modern medical diagnostics by visualizing radio-tracer distribution to reveal physiological processes. Accurate organ segmentation from PET images is essential for comprehensive multi-systemic analysis of interactions between different organs and pathologies. Existing segmentation methods are limited by insufficient annotation data and varying levels of annotation, resulting in weak generalization ability and difficulty in clinical application. Recent developments in segmentation foundation models have shown superior versatility across diverse segmentation tasks. Despite the efforts of medical adaptations, these works primarily focus on structural medical images with detailed physiological structural information and exhibit limited generalization performance on molecular PET imaging. In this paper, we collect and construct PETS-5k, the largest PET segmentation dataset to date, comprising 5,731 three-dimensional whole-body PET images and encompassing over 1.3M 2D images. Based on the established dataset, we develop SegAnyPET, a modality-specific 3D foundation model for universal promptable segmentation from PET images. To issue the challenge of discrepant annotation quality, we adopt a cross prompting confident learning (CPCL) strategy with an uncertainty-guided self-rectification process to robustly learn segmentation from high-quality labeled data and low-quality noisy labeled data for promptable segmentation. Experimental results demonstrate that SegAnyPET can segment seen and unseen target organs using only one or a few prompt points, outperforming state-of-the-art foundation models and task-specific fully supervised models with higher accuracy and strong generalization ability for universal segmentation.

Graph neural networks (GNNs) have shown considerable promise in computational chemistry. However, the limited availability of molecular data raises concerns regarding GNNs' ability to effectively capture the fundamental principles of physics and chemistry, which constrains their generalization capabilities. To address this challenge, we introduce a novel self-supervised approach termed Equivariant Masked Position Prediction (EMPP), grounded in intramolecular potential and force theory. Unlike conventional attribute masking techniques, EMPP formulates a nuanced position prediction task that is more well-defined and enhances the learning of quantum mechanical features. EMPP also bypasses the approximation of the Gaussian mixture distribution commonly used in denoising methods, allowing for more accurate acquisition of physical properties. Experimental results indicate that EMPP significantly enhances performance of advanced molecular architectures, surpassing state-of-the-art self-supervised approaches. Our code is released in https://github.com/ajy112/EMPP.

In recent years, multimodal large language models (MLLMs) with vision processing capability have shown substantial advancements, excelling particularly in interpreting general images. Their application in domain-specific tasks, like those in the medical fields, is further enhanced through continuous visual instruction fine-tuning (CVIF). Despite these advancements, a significant challenge arises from label noise encountered during the collection of domain-specific data. Our studies reveal that this label noise can adversely affect the learning of vision projection embeddings and contribute to inaccuracies in LLMs’ fine-tuning, often leading to hallucinations. In this paper, we introduce a novel framework designed to minimize the impact of label noise. Our approach focuses on stabilizing the learning of vision embeddings and reducing the effect of label noise through the inherent semantic understanding of uncertainty in LLMs. Extensive experiments demonstrate that our framework maintains robust performance in general visual question-answer (VQA) tasks while showing significant effectiveness in medical VQA tasks. To the best of our knowledge, this is the first study to specifically address and analyze the impact of label noise in CVIF.

We propose Network Automatic Relevance Determination (NARD), an extension of ARD for linearly probabilistic models, to simultaneously model sparse relationships between inputs $X \in \mathbb R^{d \times N}$ and outputs $Y \in \mathbb R^{m \times N}$, while capturing the correlation structure among the $Y$. NARD employs a matrix normal prior which contains a sparsity-inducing parameter to identify and discard irrelevant features, thereby promoting sparsity in the model. Algorithmically, it iteratively updates both the precision matrix and the relationship between $Y$ and the refined inputs. To mitigate the computational inefficiencies of the $\mathcal O(m^3 + d^3)$ cost per iteration, we introduce Sequential NARD, which evaluates features sequentially, and a Surrogate Function Method, leveraging an efficient approximation of the marginal likelihood and simplifying the calculation of determinant and inverse of an intermediate matrix. Combining the Sequential update with the Surrogate Function method further reduces computational costs. The computational complexity per iteration for these three methods is reduced to $\mathcal O(m^3+p^3)$, $\mathcal O(m^3 + d^2)$, $\mathcal O(m^3+p^2)$ respectively, where $p \ll d$ is the final number of features in the model. Our methods demonstrate significant improvements in computational efficiency with comparable performance on both synthetic and real-world datasets.

The advent of single-cell Assay for Transposase-Accessible Chromatin using sequencing (scATAC-seq) offers an innovative perspective for deciphering regulatory mechanisms by assembling a vast repository of single-cell chromatin accessibility data. While foundation models have achieved significant success in single-cell transcriptomics, there is currently no foundation model for scATAC-seq that supports zero-shot high-quality cell identification and comprehensive multi-omics analysis simultaneously. Key challenges lie in the high dimensionality and sparsity of scATAC-seq data, as well as the lack of a standardized schema for representing open chromatin regions (OCRs). Here, we present ChromFound, a foundation model tailored for scATAC-seq. ChromFound utilizes a hybrid architecture and genome-aware tokenization to effectively capture genome-wide long contexts and regulatory signals from dynamic chromatin landscapes. Pretrained on 1.97 million cells from 30 tissues and 6 disease conditions, ChromFound demonstrates broad applicability across 6 diverse tasks. Notably, it achieves robust zero-shot performance in generating universal cell representations and exhibits excellent transferability in cell type annotation and cross-omics prediction. By uncovering enhancer-gene links undetected by existing computational methods, ChromFound offers a promising framework for understanding disease risk variants in the noncoding genome. The implementation of ChromFound is available via https://github.com/JohnsonKlose/ChromFound.

Semi-supervised learning has attracted much attention due to its less dependence on acquiring abundant annotations from experts compared to fully supervised methods, which is especially important for medical image segmentation which typically requires intensive pixel/voxel-wise labeling by domain experts. Although semi-supervised methods can improve the performance by utilizing unlabeled data, there are still gaps between fully supervised methods under extremely limited annotation scenarios. In this paper, we propose a simple yet efficient strategy to explore the usage of the Segment Anything Model (SAM) for enhancing semi-supervised medical image segmentation. Concretely, the segmentation model trained with domain knowledge provides information for localization and generating input prompts to the SAM. Then the generated pseudo-labels of SAM are utilized as additional supervision to assist in the learning procedure of the semi-supervised framework. Extensive experiments demonstrate that SemiSAM significantly improves the performance of existing semi-supervised frameworks when only one or a few labeled images are available and shows strong efficiency as a plug-and-play strategy for semi-supervised medical image segmentation.

Equivariant message passing neural networks have emerged as the prevailing approach for predicting chemical properties of molecules due to their ability to leverage translation and rotation symmetries, resulting in a strong inductive bias. However, the equivariant operations in each layer can impose excessive constraints on the function form and network flexibility. To address these challenges, we introduce a novel network called the Hybrid Directional Graph Neural Network (HDGNN), which effectively combines strictly equivariant operations with learnable modules. We evaluate the performance of HDGNN on the QM9 dataset and the IS2RE dataset of OC20, demonstrating its state-of-the-art performance on several tasks and competitive performance on others. Our code is anonymously released on https://github.com/ajy112/HDGNN.

Graph Neural Networks (GNNs) have gained significant attention as a powerful modeling and inference method, especially for homophilic graph-structured data. To empower GNNs in heterophilic graphs, where adjacent nodes exhibit dissimilar labels or features, Signed Message Passing (SMP) has been widely adopted. However, there is a lack of theoretical and empirical analysis regarding the limitations of SMP. In this work, we unveil the potential pitfalls of SMP and their remedies. We first identify two limitations of SMP: undesirable representation update for multi-hop neighbors and vulnerability against oversmoothing issues. To overcome these challenges, we propose a novel message-passing function called Multiset to Multiset GNN (M2M-GNN). Our theoretical analyses and extensive experiments demonstrate that M2M-GNN effectively alleviates the limitations of SMP, yielding superior performance in comparison.

Recently, significant advancements have been made in Large Language Models (LLMs) through the implementation of various alignment techniques. These techniques enable LLMs to generate highly tailored content in response to diverse user instructions. Consequently, LLMs have the potential to serve as robust, customizable recommendation systems in the field of content recommendation. However, using LLMs with user individual information and online exploration remains a challenge, which are important perspectives in developing personalized news headline generation algorithms. In this paper, we propose a novel framework to generate personalized news headlines using LLMs with extensive online exploration. The proposed approach involves initially training an offline goal-conditioned policy using supervised learning. Subsequently, online exploration is employed to collect new data for the next training iteration. Results from simulations, experiments, and real-word scenario demonstrate that our framework achieves outstanding performance on established benchmarks and can effectively generate personalized headlines under different reward settings. By treating the LLM as a goal-conditioned agent, the model can perform online exploration by modifying the goals without frequently retraining the model. To the best of our knowledge, this work represents the first investigation into the capability of LLMs to generate customized news headlines with goal-conditioned reinforcement learning via supervised learning within LLMs.

In recent advancements, large language models (LLMs) have demonstrated remarkable capabilities in diverse tasks, primarily through interactive question-answering with humans. This development marks significant progress towards artificial general intelligence (AGI). Despite their superior performance, LLMs often exhibit limitations when adapted to domain-specific tasks through instruction fine-tuning (IF). The primary challenge lies in the discrepancy between the data distribution in general and domain-specific contexts, leading to suboptimal accuracy in specialized tasks. To address this, continual instruction fine-tuning (CIF), particularly supervised fine-tuning (SFT), on targeted domain-specific instruction datasets is necessary. Our ablation study reveals that the structure of these instruction datasets critically influences CIF performance, with substantial data distributional shifts resulting in notable performance degradation. In this paper, we introduce a novel framework that enhances CIF by promoting format uniformity. We assess our approach using the Llama2 chat model across various domain-specific instruction datasets. The results demonstrate not only an improvement in task-specific performance under CIF but also a reduction in catastrophic forgetting (CF). This study contributes to the optimization of LLMs for domain-specific applications, highlighting the significance of data structure and distribution in CIF.

Graph Neural Networks (GNNs) have attracted much attention due to their superior learning capability. Despite the successful applications of GNNs in many areas, their performance suffers heavily from the long-tailed node degree distribution. Most prior studies tackle this issue by devising sophisticated model architectures. In this article, we aim to improve the performance of tail nodes (low-degree or hard-to-classify nodes) via a generic and light normalization method. In detail, we propose a novel normalization method for GNNs, termed as ResNorm, whichReshapes a long-tailed distribution into a normal-like distribution viaNormalization. The ResNorm includes two operators. First, thescaleoperator reshapes the distribution of the node-wise standard deviation (NStd) so as to improve the accuracy of tail nodes. Second, the analysis of the behavior of the standard shift indicates that the standard shift serves as a preconditioner on the weight matrix, increasing the risk of over-smoothing. To address this issue, we design a newshiftoperator for ResNorm, which simulates the degree-specific parameter strategy in a low-cost manner. Extensive experiments on various node classification benchmark datasets have validated the effectiveness of ResNorm in improving the performance of tail nodes as well as the overall performance.

Online platforms in the Internet Economy commonly incorporate recommender systems that recommend products (or “arms”) to users (or “agents”). A key challenge in this domain arises from myopic agents who are naturally incentivized to exploit by choosing the optimal arm based on current information, rather than exploring various alternatives to gather information that benefits the collective. We propose a new recommender system that aligns with agents’ incentives while achieving asymptotically optimal performance, as measured by regret in repeated interactions. Our framework models this incentive-aware system as a multi-agent bandit problem in two-sided markets, where the interactions of agents and arms are facilitated by recommender systems on online platforms. This model incorporates incentive constraints induced by agents’ opportunity costs. In scenarios where opportunity costs are known to the platform, we show the existence of an incentive-compatible recommendation algorithm. This algorithm pools recommendations between a genuinely good arm and an unknown arm using a randomized and adaptive strategy. Moreover, when these opportunity costs are unknown, we introduce an algorithm that randomly pools recommendations across all arms, utilizing the cumulative loss from each arm as feedback for strategic exploration. We demonstrate that both algorithms satisfy an ex-post fairness criterion, which protects agents from over-exploitation. All code for using the proposed algorithms and reproducing results is made available on GitHub.

Typical machine learning applications always assume the data follows independent and identically distributed (IID) assumptions. In contrast, this assumption is frequently violated in real-world circumstances, leading to the Out-of-Distribution (OOD) generalization problem and a major drop in model robustness. To mitigate this issue, the invariant learning technique is leveraged to distinguish between spurious features and invariant features among all input features and to train the model purely on the basis of the invariant features. Numerous invariant learning strategies imply that the training data should contain domain information. Such information includes the environment index or auxiliary information acquired from prior knowledge. However, acquiring these information is typically impossible in practice. In this study, we present TIVA for environment-independent invariance learning, which requires no environment-specific information in training data. We discover and prove that, given certain mild data conditions, it is possible to train an environment partitioning policy based on attributes that are independent of the targets and then conduct invariant risk minimization. We examine our method in comparison to other baseline methods, which demonstrate superior performance and excellent robustness under OOD, using multiple benchmarks.

In recent years, the explosion of web videos makes text-video retrieval increasingly essential and popular for video filtering, recommendation, and search. Text-video retrieval aims to rank relevant text/video higher than irrelevant ones. The core of this task is to precisely measure the cross-modal similarity between texts and videos. Recently, contrastive learning methods have shown promising results for text-video retrieval, most of which focus on the construction of positive and negative pairs to learn text and video representations. Nevertheless, they do not pay enough attention to hard negative pairs and lack the ability to model different levels of semantic similarity. To address these two issues, this paper improves contrastive learning using two novel techniques. First, to exploit hard examples for robust discriminative power, we propose a novel Dual-Modal Attention-Enhanced Module (DMAE) to mine hard negative pairs from textual and visual clues. By further introducing a Negative-aware InfoNCE (NegNCE) loss, we are able to adaptively identify all these hard negatives and explicitly highlight their impacts in the training loss. Second, our work argues that triplet samples can better model fine-grained semantic similarity compared to pairwise samples. We thereby present a new Triplet Partial Margin Contrastive Learning (TPM-CL) module to construct partial order triplet samples by automatically generating fine-grained hard negatives for matched text-video pairs. The proposed TPM-CL designs an adaptive token masking strategy with cross-modal interaction to model subtle semantic differences. Extensive experiments demonstrate that the proposed approach outperforms existing methods on four widely-used text-video retrieval datasets, including MSR-VTT, MSVD, DiDeMo and ActivityNet.

Deep sequence networks such as multi-head self-attention networks provide a promising way to extract effective representations from raw sequence data in an end-to-end fashion and have shown great success in various domains such as natural language processing, computer vision, $etc$. However, in domains such as financial risk management and anti-fraud where expert-derived features are heavily relied on, deep sequence models struggle to dominate the game.In this paper, we introduce a simple framework called symbolic testing to verify the learnability of certain expert-derived features over sequence data. A systematic investigation over simulated data reveals the fact that the self-attention architecture fails to learn some standard symbolic expressions like the count distinct operation. To overcome this deficiency, we propose a novel architecture named SHORING, which contains two components:event network andsequence network. Theevent network efficiently learns arbitrary high-orderevent-level conditional embeddings via a reparameterization trick while thesequence network integrates domain-specific aggregations into the sequence-level representation, thereby providing richer inductive biases compare to standard sequence architectures like self-attention. We conduct comprehensive experiments and ablation studies on synthetic datasets that mimic sequence data commonly seen in anti-fraud domain and three real-world datasets. The results show that SHORING learns commonly used symbolic features well, and experimentally outperforms the state-of-the-art methods by a significant margin over real-world online transaction datasets. The symbolic testing framework and SHORING have been applied in anti-fraud model development at Alipay and improved performance of models for real-time fraud-detection.

With the explosive growth of e-payment industry, online transaction fraud has become one of the biggest challenges for the business. The historical behavior information of users provides rich information for digging into the users' fraud risk. While considerable efforts have been made in this direction, a long-standing challenge is how to effectively exploit user's behavioral information and provide explainable prediction results. In fact, the value variations of same field from different events and the interactions of different fields within one event have proven to be strong indicators of fraudulent behaviors. In this paper, we propose the Dual Importance-aware Factorization Machines (DIFM), which exploits the inter- and intra-event information among users’ behavior sequence from dual perspectives, i.e., field value variations and field interactions simultaneously for fraud detection. The proposed model is deployed in Alipay's risk management system, which provides real-time fraud detection service for e-commerce platforms. Experimental results on industrial data under various scenarios in the platform clearly demonstrate that our model achieves significant improvements compared with various state-of-the-art baseline models. Moreover, the \model~could also give an insight into the explanation of the prediction results from dual perspectives.

Knowledge graph is generally incorporated into recommender systems to improve overall performance. Due to the generalization and scale of the knowledge graph, most knowledge relationships are not helpful for a target user-item prediction. To exploit the knowledge graph to capture target-specific knowledge relationships in recommender systems, we need to distill the knowledge graph to reserve the useful information and refine the knowledge to capture the users' preferences. To address the issues, we propose Knowledge-aware Conditional Attention Networks (KCAN), which is an end-to-end model to incorporate knowledge graph into a recommender system. Specifically, we use a knowledge-aware attention propagation manner to obtain the node representation first, which captures the global semantic similarity on the user-item network and the knowledge graph. Then given a target, i.e., a user-item pair, we automatically distill the knowledge graph into the target-specific subgraph based on the knowledge-aware attention. Afterward, by applying a conditional attention aggregation on the subgraph, we refine the knowledge graph to obtain target-specific node representations. Therefore, we can gain both representability and personalization to achieve overall performance. Experimental results on real-world datasets demonstrate the effectiveness of our framework over the state-of-the-art algorithms.

Credit risk prediction is a fundamental problem for most financial institutions.Previous methods mainly adopt users' individual features on a single snapshot.However, users' individual features on financial platforms are usually too sparse to be informative.And previous methods ignore that the features, the behaviours and the credit risk of the users are all dynamic.To resolve the problems, we aim to model the credit risk prediction on dynamic graphs and propose a Temporal-Aware Graph Neural Network (TemGNN) to predict user credit risk.In detail, the model consists of three parts: i) a static model to extract the user's static factors regarding the credit risk.ii) a short-term graph encoder with special graph convolution modules for each snapshot to enrich the user's information through aggregating short-term temporal and structural information.iii) a long-term temporal model based on LSTM with interval-decayed attention to adaptively aggregate the long-term information from the static factors and interval-irregular dynamic snapshots.By combining the three parts together, our model is able to mine both the short-and long-term temporal-structural information.Experimentally, we use the users' authorized lending behaviours as the temporal graphs to do default prediction on Alipay.The results show that our model achieves the best performance among the state-of-the-art methods.

In the past decade, there has been substantial progress at training increasingly deep neural networks. Recent advances within the teacher–student training paradigm have established that information about past training updates show promise as a source of guidance during subsequent training steps. Based on this notion, in this paper, we propose Long Short-Term Sample Distillation, a novel training policy that simultaneously leverages multiple phases of the previous training process to guide the later training updates to a neural network, while efficiently proceeding in just one single generation pass. With Long Short-Term Sample Distillation, the supervision signal for each sample is decomposed into two parts: a long-term signal and a short-term one. The long-term teacher draws on snapshots from several epochs ago in order to provide steadfast guidance and to guarantee teacher–student differences, while the short-term one yields more up-to-date cues with the goal of enabling higher-quality updates. Moreover, the teachers for each sample are unique, such that, overall, the model learns from a very diverse set of teachers. Comprehensive experimental results across a range of vision and NLP tasks demonstrate the effectiveness of this new training method.

We address a practical problem ubiquitous in modern marketing campaigns, in which a central agent tries to learn a policy for allocating strategic financial incentives to customers and observes only bandit feedback. In contrast to traditional policy optimization frameworks, we take into account the additional reward structure and budget constraints common in this setting, and develop a new two-step method for solving this constrained counterfactual policy optimization problem. Our method first casts the reward estimation problem as a domain adaptation problem with supplementary structure, and then subsequently uses the estimators for optimizing the policy with constraints. We also establish theoretical error bounds for our estimation procedure and we empirically show that the approach leads to significant improvement on both synthetic and real datasets.

Chinese Spelling Check (CSC) is a task to detect and correct spelling errors in Chinese natural language.Existing methods have made attempts to incorporate the similarity knowledge between Chinese characters.However, they take the similarity knowledge as either an external input resource or just heuristic rules.This paper proposes to incorporate phonological and visual similarity knowledge into language models for CSC via a specialized graph convolutional network (SpellGCN).The model builds a graph over the characters, and SpellGCN is learned to map this graph into a set of inter-dependent character classifiers.These classifiers are applied to the representations extracted by another network, such as BERT, enabling the whole network to be end-to-end trainable.Experiments 1 are conducted on three human-annotated datasets.Our method achieves superior performance against previous models by a large margin.

Dynamic graphs such as the user-item interactions graphs and financial transaction networks are ubiquitous nowadays. While numerous representation learning methods for static graphs have been proposed, the study of dynamic graphs is still in its infancy. A main challenge of modeling dynamic graphs is how to effectively encode temporal and structural information into nonlinear and compact dynamic embeddings. To achieve this, we propose a principled graph-neural-based approach to learn continuous-time dynamic embeddings. We first define a temporal dependency interaction graph(TDIG) that is induced from sequences of interaction data. Based on the topology of this TDIG, we develop a dynamic message passing neural network named TDIG-MPNN, which can capture the fine-grained global and local information on TDIG. In addition, to enhance the quality of continuous-time dynamic embeddings, a novel selection mechanism comprised of two successive steps, i.e., co-attention and gating, is applied before the above TDIG-MPNN layer to adjust the importance of the nodes by considering high-order correlation between interactive nodes' k-depth neighbors on TDIG. Finally, we cast our learning problem in the framework of temporal point processes (TPPs) where we use TDIG-MPNN to design a neural intensity function for the dynamic interaction processes. Our model achieves superior performance over alternatives on temporal interaction prediction (including tranductive and inductive tasks) on multiple datasets.

Aiming to effectively distinguish loan default in the Mobile Credit Payment Service, industrial efforts mainly attempt to employ conventional classifier with complicated feature engineer for prediction. However, these solutions fail to exploit multiplex relations existed in the financial scenarios and ignore the key intrinsic properties of the loan default detection, i.e., communicability, complementation and induction. To address these issues, we develop a novel attributed multiplex graph based loan default detection approach for effectively integrating multiplex relations in financial scenarios. Considering the complexity of financial scenario, an Attributed Multiplex Graph (AMG) is proposed to jointly model various relations and objects as well as the rich attributes on nodes and edges. We elaborately design relation-specific receptive layers equipped with adaptive breadth function to incorporate important information derived from local structure in each aspect of AMG and stack multiple propagation layer to explore the high-order connectivity information. Furthermore, a relation-specific attention mechanism is adopted to emphasize relevant information during end-to-end training. Extensive experiments conducted on the large-scale real- world dataset verify the effectiveness of the proposed model com- pared with state of arts. Moreover, AMG-DP has also achieved a performance improvement of 9.37% on KS metric in recent months after successful deployment in the Alipay APP.

Today text classification models have been widely used. However, these classifiers are found to be easily fooled by adversarial examples. Fortunately, standard attacking methods generate adversarial texts in a pair-wise way, that is, an adversarial text can only be created from a real-world text by replacing a few words. In many applications, these texts are limited in numbers, therefore their corresponding adversarial examples are often not diverse enough and sometimes hard to read, thus can be easily detected by humans and cannot create chaos at a large scale. In this paper, we propose an end to end solution to efficiently generate adversarial texts from scratch using generative models, which are not restricted to perturbing the given texts. We call it unrestricted adversarial text generation. Specifically, we train a conditional variational autoencoder (VAE) with an additional adversarial loss to guide the generation of adversarial examples. Moreover, to improve the validity of adversarial texts, we utilize discrimators and the training framework of generative adversarial networks (GANs) to make adversarial texts consistent with real data. Experimental results on sentiment analysis demonstrate the scalability and efficiency of our method. It can attack text classification models with a higher success rate than existing methods, and provide acceptable quality for humans in the meantime.

Kunlong Chen, Weidi Xu, Xingyi Cheng, Zou Xiaochuan, Yuyu Zhang, Le Song, Taifeng Wang, Yuan Qi, Wei Chu. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020.

We propose a modeling framework for event data and aim to answer questions such as \emph{when} and \emph{why} the next event would happen. Our proposed model excels in small data regime with the ability to incorporate domain knowledge in terms of logic rules. We model the dynamics of the event starts and ends via intensity function with the structures informed by a set of first-order temporal logic rules. Using the softened representation of temporal relations, and a weighted combination of logic rules, our probabilistic model can deal with uncertainty in events. Furthermore, many well-known point processes (e.g., Hawkes process, self-correcting point process) can be interpreted as special cases of our model given simple temporal logic rules. Our model, therefore, riches the family of point processes. We derive a maximum likelihood estimation procedure for our model and show that it can lead to accurate predictions when data are sparse and domain knowledge is critical.

Small and Medium-sized Enterprises (SMEs) are playing a vital role in the modern economy. Recent years, financial risk analysis for SMEs attracts lots of attentions from financial institutions. However, the financial risk analysis for SMEs usually suffers data deficiency problem, especially for the mobile financial institutions which seldom collect credit-related data directly from SMEs. Fortunately, although credit-related information of SMEs is hard to be acquired sufficiently, the interactive relationships between SMEs, which may contain valuable information of financial risk, is usually available for the mobile financial institutions. Finding out credit-related relationship of SME from massive interactions helps comprehensively model the SMEs thus improve the performance of financial risk analysis. In this paper, tackling the data deficiency problem of financial risk analysis for SMEs, we propose an innovative financial risk analysis framework with graph-based supply chain mining. Specifically, to capture the credit-related topology structural and temporal variation information of SMEs, we design and employ a novel spatial-temporal aware graph neural network, to mine supply chain relationship on a SME graph, and then analysis the credit risk based on the mined supply chain graph. Experimental results on real-world financial datasets prove the effectiveness of our proposal for financial risk analysis for SMEs.

Recently, in order to take a preemptive opportunity in the mobile economy, the Internet companies conduct thousands of marketing campaigns every day, to promote their mobile products and services. In the mobile marketing scenario, one of the fundamental issues is the audience expansion task for marketing campaigns. Given a set of seed users, audience expansion aims to seek more users (audiences), who are similar to the seeds and will finish the business goal of the targeted campaign (ie convert). However, the problem is challenging in three aspects. First, a company will run hundreds of campaigns to serve massive users every day. The requirements of scalability and timeliness make training model for each campaign extremely resource-consuming thus impractical. Therefore, we proposed to solve the problem in a two-stage manner, in which the offline stage employs heavyweight user representation learning and the online stage performs embedding-based lightweight audience expansion. Second, conventional two-stage audience expansion systems neglect the high-order user-campaign interactions and usually generate entangled user embeddings, thus fail to achieve high-quality user representation. Third, the seeds, which are usually provided by experts or collected from users' feedbacks, could be noisy and cannot cover the entire actual audiences, thus introduce coverage bias. Unfortunately, to our best knowledge, none of the related literatures tackle this crucial issue of audience expansion.

Several sampling algorithms with variance reduction have been proposed for accelerating the training of Graph Convolution Networks (GCNs). However, due to the intractable computation of optimal sampling distribution, these sampling algorithms are suboptimal for GCNs and are not applicable to more general graph neural networks (GNNs) where the message aggregator contains learned weights rather than fixed weights, such as Graph Attention Networks (GAT). The fundamental reason is that the embeddings of the neighbors or learned weights involved in the optimal sampling distribution are \emph{changing} during the training and \emph{not known a priori}, but only \emph{partially observed} when sampled, thus making the derivation of an optimal variance reduced samplers non-trivial. In this paper, we formulate the optimization of the sampling variance as an adversary bandit problem, where the rewards are related to the node embeddings and learned weights, and can vary constantly. Thus a good sampler needs to acquire variance information about more neighbors (exploration) while at the same time optimizing the immediate sampling variance (exploit). We theoretically show that our algorithm asymptotically approaches the optimal variance within a factor of 3. We show the efficiency and effectiveness of our approach on multiple datasets.

Machine learning over graphs has been emerging as powerful learning tools for graph data. However, it is challenging for industrial communities to leverage the techniques, such as graph neural networks (GNNs), and solve real-world problems at scale because of inherent data dependency in the graphs. As such, we cannot simply train a GNN with classic learning systems, for instance, parameter server that assumes data parallelism. Existing systems store the graph data in-memory for fast accesses either in a single machine or graph stores from remote. The major drawbacks are three-fold. First, they cannot scale because of the limitations on the volume of the memories, or the bandwidth between graph stores and workers. Second, they require extra development of graph stores without well exploiting mature infrastructures such as MapReduce that guarantee good system properties. Third, they focus on training but ignore optimizing the performance of inference over graphs, thus makes them an unintegrated system. In this paper, we design AGL, a scalable and integrated system, with fully-functional training and inference for GNNs. Our system design follows the message passing scheme underlying the computations of GNNs. We design to generate the K -hop neighborhood, an information-complete subgraph for each node, as well as do the inference simply by merging values from in-edge neighbors and propagating values to out-edge neighbors via MapReduce. In addition, the K -hop neighborhood contains information-complete subgraphs for each node, thus we simply do the training on parameter servers due to data independence. Our system AGL, implemented on mature infrastructures, can finish the training of a 2-layer GNN on a graph with billions of nodes and hundred billions of edges in 14 hours, and complete the inference in 1.2 hours.

Point-of-Interest (POI) recommendation has been extensively studied and successfully applied in industry recently. However, most existing approaches build centralized models on the basis of collecting users’ data. Both private data and models are held by the recommender, which causes serious privacy concerns. In this article, we propose a novel Privacy preserving POI Recommendation (PriRec) framework. First, to protect data privacy, users’ private data (features and actions) are kept on their own side, e.g., Cellphone or Pad. Meanwhile, the public data that need to be accessed by all the users are kept by the recommender to reduce the storage costs of users’ devices. Those public data include: (1) static data only related to the status of POI, such as POI categories, and (2) dynamic data dependent on user-POI actions such as visited counts. The dynamic data could be sensitive, and we develop local differential privacy techniques to release such data to the public with privacy guarantees. Second, PriRec follows the representations of Factorization Machine (FM) that consists of a linear model and the feature interaction model. To protect the model privacy, the linear models are saved on the users’ side, and we propose a secure decentralized gradient descent protocol for users to learn it collaboratively. The feature interaction model is kept by the recommender since there is no privacy risk, and we adopt a secure aggregation strategy in a federated learning paradigm to learn it. To this end, PriRec keeps users’ private raw data and models in users’ own hands, and protects user privacy to a large extent. We apply PriRec in real-world datasets, and comprehensive experiments demonstrate that, compared with FM, PriRec achieves comparable or even better recommendation accuracy.

As one of the major frauds in financial services, cash-out fraud is that users pursue cash gains with illegal or insincere means. Conventional solutions for the cash-out user detection are to perform subtle feature engineering for each user and then apply a classifier, such as GDBT and Neural Network. However, users in financial services have rich interaction relations, which are seldom fully exploited by conventional solutions. In this paper, with the real datasets in Ant Credit Pay of Ant Financial Services Group, we first study the cashout user detection problem and propose a novel hierarchical attention mechanism based cash-out user detection model, called HACUD. Specifically, we model different types of objects and their rich attributes and interaction relations in the scenario of credit payment service with an Attributed Heterogeneous Information Network (AHIN). The HACUD model enhances feature representation of objects through meta-path based neighbors exploiting different aspects of structure information in AHIN. Furthermore, a hierarchical attention mechanism is elaborately designed to model user’s preferences towards attributes and meta-paths. Experimental results on two real datasets show that the HACUD outperforms the state-of-the-art methods.

Current Internet market makers are facing an intense competitive environment, where personalized price reductions or discounted coupons are provided by their peers to attract more customers. Much investment is spent to catch up with each other’s competitors but participants in such a price cut war are often incapable of winning due to their lack of information about others’ strategies or customers’ preference. We formalize the problem as a stochastic game with imperfect and incomplete information and develop a variant of Latent Dirichlet Allocation (LDA) to infer latent variables under the current market environment, which represents preferences of customers and strategies of competitors. Tests on simulated experiments and an open dataset for real data show that, by subsuming all available market information of the market maker’s competitors, our model exhibits a significant improvement for understanding the market environment and finding the best response strategies in the Internet price war. Our work marks the first successful learning method to infer latent information in the environment of price war by the LDA modeling, and sets an example for related competitive applications to follow.

We present, GeniePath, a scalable approach for learning adaptive receptive fields of neural networks defined on permutation invariant graph data. In GeniePath, we propose an adaptive path layer consists of two complementary functions designed for breadth and depth exploration respectively, where the former learns the importance of different sized neighborhoods, while the latter extracts and filters signals aggregated from neighbors of different hops away. Our method works in both transductive and inductive settings, and extensive experiments compared with competitive methods show that our approaches yield state-of-the-art results on large graphs.

The insurance industry has been creating innovative products around the emerging online shopping activities. Such ecommerce insurance is designed to protect buyers from potential risks such as impulse purchases and counterfeits. Fraudulent claims towards online insurance typically involve multiple parties such as buyers, sellers, and express companies, and they could lead to heavy financial losses. In order to uncover the relations behind organized fraudsters and detect fraudulent claims, we developed a large-scale insurance fraud detection system, i.e., InfDetect, which provides interfaces for commonly used graphs, standard data processing procedures, and a uniform graph learning platform. InfDetect is able to process big graphs containing up to 100 millions of nodes and billions of edges.In this paper, we investigate different graphs to facilitate fraudster mining, such as a device-sharing graph, a transaction graph, a friendship graph, and a buyer-seller graph. These graphs are fed to a uniform graph learning platform containing supervised and unsupervised graph learning algorithms. Cases on widely applied e-commerce insurance are described to demonstrate the usage and capability of our system. InfDetect has successfully detected thousands of fraudulent claims and saved over tens of thousands of dollars daily.

Mobile payment such as Alipay has been widely used in our daily lives. To further promote the mobile payment activities, it is important to run marketing campaigns under a limited budget by providing incentives such as coupons, commissions to merchants. As a result, incentive optimization is the key to maximizing the commercial objective of the marketing campaign. With the analyses of online experiments, we found that the transaction network can subtly describe the similarity of merchants' responses to different incentives, which is of great use in the incentive optimization problem. In this paper, we present a graph representation learning method atop of transaction networks for merchant incentive optimization in mobile payment marketing. With limited samples collected from online experiments, our end-to-end method first learns merchant representations based on an attributed transaction networks, then effectively models the correlations between the commercial objectives each merchant may achieve and the incentives under varying treatments. Thus we are able to model the sensitivity to incentive for each merchant, and spend the most budgets on those merchants that show strong sensitivities in the marketing campaign. Extensive offline and online experimental results at Alipay demonstrate the effectiveness of our proposed approach.

With the rapid growth of financial services, fraud detection has been a very important problem to guarantee a healthy environment for both users and providers. Conventional solutions for fraud detection mainly use some rule-based methods or distract some features manually to perform prediction. However, in financial services, users have rich interactions and they themselves always show multifaceted information. These data form a large multiview network, which is not fully exploited by conventional methods. Additionally, among the network, only very few of the users are labelled, which also poses a great challenge for only utilizing labeled data to achieve a satisfied performance on fraud detection. To address the problem, we expand the labeled data through their social relations to get the unlabeled data and propose a semi-supervised attentive graph neural network, named SemiGNN to utilize the multi-view labeled and unlabeled data for fraud detection. Moreover, we propose a hierarchical attention mechanism to better correlate different neighbors and different views. Simultaneously, the attention mechanism can make the model interpretable and tell what are the important factors for the fraud and why the users are predicted as fraud. Experimentally, we conduct the prediction task on the users of Alipay, one of the largest third-party online and offline cashless payment platform serving more than 4 hundreds of million users in China. By utilizing the social relations and the user attributes, our method can achieve a better accuracy compared with the state-of-the-art methods on two tasks. Moreover, the interpretable results also give interesting intuitions regarding the tasks.

There are great interests as well as many challenges in applying reinforcement learning (RL) to recommendation systems. In this setting, an online user is the environment; neither the reward function nor the environment dynamics are clearly defined, making the application of RL challenging. In this paper, we propose a novel model-based reinforcement learning framework for recommendation systems, where we develop a generative adversarial network to imitate user behavior dynamics and learn her reward function. Using this user model as the simulation environment, we develop a novel Cascading DQN algorithm to obtain a combinatorial recommendation policy which can handle a large number of candidate items efficiently. In our experiments with real data, we show this generative adversarial user model can better explain user behavior than alternatives, and the RL policy based on this model can lead to a better long-term reward for the user and higher click rate for the system.

We consider the networked multi-agent reinforcement learning (MARL) problem in a fully decentralized setting, where agents learn to coordinate to achieve joint success. This problem is widely encountered in many areas including traffic control, distributed control, and smart grids. We assume each agent is located at a node of a communication network and can exchange information only with its neighbors. Using softmax temporal consistency, we derive a primal-dual decentralized optimization method and obtain a principled and data-efficient iterative algorithm named {\em value propagation}. We prove a non-asymptotic convergence rate of $\mathcal{O}(1/T)$ with nonlinear function approximation. To the best of our knowledge, it is the first MARL algorithm with a convergence guarantee in the control, off-policy, non-linear function approximation, fully decentralized setting.

Fraudulent claim detection is one of the greatest challenges the insurance industry faces. Alibaba's return-freight insurance, providing return-shipping postage compensations over product return on the e-commerce platform, receives thousands of potentially fraudulent claims everyday. Such deliberate abuse of the insurance policy could lead to heavy financial losses. In order to detect and prevent fraudulent insurance claims, we developed a novel data-driven procedure to identify groups of organized fraudsters, one of the major contributions to financial losses, by learning network information. In this paper, we introduce a device-sharing network among claimants, followed by developing an automated solution for fraud detection based on graph learning algorithms, to separate fraudsters from regular customers and uncover groups of organized fraudsters. This solution applied at Alibaba achieves more than 80% precision while covering 44% more suspicious accounts compared with a previously deployed rule-based classifier after human expert investigations. Our approach can easily and effectively generalizes to other types of insurance.

With the explosive growth of e-commerce and the booming of e-payment, detecting online transaction fraud in real time has become increasingly important to Fintech business. To tackle this problem, we introduce the TitAnt, a transaction fraud detection system deployed in Ant Financial, one of the largest Fintech companies in the world. The system is able to predict online real-time transaction fraud in mere milliseconds. We present the problem definition, feature extraction, detection methods, implementation and deployment of the system, as well as empirical effectiveness. Extensive experiments have been conducted on large real-world transaction data to show the effectiveness and the efficiency of the proposed system.

Internet companies are facing the need for handling large-scale machine learning applications on a daily basis and distributed implementation of machine learning algorithms which can handle extra-large-scale tasks with great performance is widely needed. Deep forest is a recently proposed deep learning framework which uses tree ensembles as its building blocks and it has achieved highly competitive results on various domains of tasks. However, it has not been tested on extremely large-scale tasks. In this work, based on our parameter server system, we developed the distributed version of deep forest. To meet the need for real-world tasks, many improvements are introduced to the original deep forest model, including MART (Multiple Additive Regression Tree) as base learners for efficiency and effectiveness consideration, the cost-based method for handling prevalent class-imbalanced data, MART based feature selection for high dimension data, and different evaluation metrics for automatically determining the cascade level. We tested the deep forest model on an extra-large-scale task, i.e., automatic detection of cash-out fraud, with more than 100 million training samples. Experimental results showed that the deep forest model has the best performance according to the evaluation metrics from different perspectives even with very little effort for parameter tuning. This model can block fraud transactions in a large amount of money each day. Even compared with the best-deployed model, the deep forest model can additionally bring a significant decrease in economic loss each day.

Ant Credit Pay is a consumer credit service in Ant Financial Service Group. Similar to credit card, loan default is one of the major risks of this credit product. Hence, effective algorithm for default prediction is the key to losses reduction and profits increment for the company. However, the challenges facing in our scenario are different from those in conventional credit card service. The first one is scalability. The huge volume of users and their behaviors in Ant Financial requires the ability to process industrial-scale data and perform model training efficiently. The second challenges is the cold-start problem. Different from the manual review for credit card application in conventional banks, the credit limit of Ant Credit Pay is automatically offered to users based on the knowledge learned from big data. However, default prediction for new users is suffered from lack of enough credit behaviors. It requires that the proposal should leverage other new data source to alleviate the cold-start problem. Considering the above challenges and the special scenario in Ant Financial, we try to incorporate default prediction with network information to alleviate the cold-start problem. In this paper, we propose an industrial-scale distributed network representation framework, termed NetDP, for default prediction in Ant Credit Pay. The proposal explores network information generated by various interaction between users, and blends unsupervised and supervised network representation in a unified framework for default prediction problem. Moreover, we present a parameter-server-based distributed implement of our proposal to handle the scalability challenge. Experimental results demonstrate the effectiveness of our proposal, especially in cold-start problem, as well as the efficiency for industrial-scale dataset.

With the rise of the Internet industry and the technique of artificial intelligence, personalized services are increasingly important in recent years for improving user experience and increasing corporates' competitiveness and profits. Precise prediction of customers' behaviors has shown great effects in modern business marketing, especially when making personalized decisions. In this paper, we develop a deep learning network to make personalized predictions of their behaviors among a list of potential choices. The architecture of this model combines each user's features and his historical event lists by sequence-to-sequence (Seq2Seq) structure and make predictions based on his recent event lists. We also modify the long-short- term memory (LSTM) cell forget gate's structure to enhance the attention ability. Such design, called the attetioned LSTM, converges quicker and better while still maintain the similar performance in open dataset IMDB. In addition, in dealing with personalized prediction problems in real-world datasets provided by our cooperative company, our attentioned LSTM achieves a 10% higher precision in average than the standard LSTM model. The advantage is confirmed in evaluation of this generic method on a real dataset of users' behaviors sequences and individuals' attribute profiles from Ant Financial. It also achieves a great result working on the real-world business scene. This model can also achieve a great performance working on the real-world business scene.

In this work, we propose to actively detect implicit fraudulent transactions. A novel machine learning method is introduced to distinguish anomalous electronic transactions based on the historical records. The transferor will be alerted during the on-going payment when the fraud probability is recognized as large enough. Compared with elaborative rule-based approaches, our model is much more effective in fraud detection.

Gaussian processes (GPs) are powerful non-parametric function estimators. However, their applications are largely limited by the expensive computational cost of the inference procedures. Existing stochastic or distributed synchronous variational inferences, although have alleviated this issue by scaling up GPs to millions of samples, are still far from satisfactory for real-world large applications, where the data sizes are often orders of magnitudes larger, say, billions. To solve this problem, we propose ADVGP, the first Asynchronous Distributed Variational Gaussian Process inference for regression, on the recent large-scale machine learning platform, PARAMETER SERVER. ADVGP uses a novel, flexible variational framework based on a weight space augmentation, and implements the highly efficient, asynchronous proximal gradient optimization. While maintaining comparable or better predictive performance, ADVGP greatly improves upon the efficiency of the existing variational methods. With ADVGP, we effortlessly scale up GP regression to a real-world application with billions of samples and demonstrate an excellent, superior prediction accuracy to the popular linear models.

In recent years, due to the emergence of Big Data (terabytes or petabytes) and Big Model (tens of billions of parameters), there has been an ever-increasing need of parallelizing machine learning (ML) algorithms in both academia and industry. Although there are some existing distributed computing systems, such as Hadoop and Spark, for parallelizing ML algorithms, they only provide synchronous and coarse-grained operators (e.g., Map, Reduce, and Join, etc.), which may hinder developers from implementing more efficient algorithms. This motivated us to design a universal distributed platform termed KunPeng, that combines both distributed systems and parallel optimization algorithms to deal with the complexities that arise from large-scale ML. Specifically, KunPeng not only encapsulates the characteristics of data/model parallelism, load balancing, model sync-up, sparse representation, industrial fault-tolerance, etc., but also provides easy-to-use interface to empower users to focus on the core ML logics. Empirical results on terabytes of real datasets with billions of samples and features demonstrate that, such a design brings compelling performance improvements on ML programs ranging from Follow-the-Regularized-Leader Proximal algorithm to Sparse Logistic Regression and Multiple Additive Regression Trees. Furthermore, KunPeng's encouraging performance is also shown for several real-world applications including the Alibaba's Double 11 Online Shopping Festival and Ant Financial's transaction risk estimation.

MOTIVATION: High-throughput mRNA sequencing (RNA-Seq) is a powerful tool for quantifying gene expression. Identification of transcript isoforms that are differentially expressed in different conditions, such as in patients and healthy subjects, can provide insights into the molecular basis of diseases. Current transcript quantification approaches, however, do not take advantage of the shared information in the biological replicates, potentially decreasing sensitivity and accuracy. RESULTS: We present a novel hierarchical Bayesian model called Differentially Expressed Isoform detection from Multiple biological replicates (DEIsoM) for identifying differentially expressed (DE) isoforms from multiple biological replicates representing two conditions, e.g. multiple samples from healthy and diseased subjects. DEIsoM first estimates isoform expression within each condition by (1) capturing common patterns from sample replicates while allowing individual differences, and (2) modeling the uncertainty introduced by ambiguous read mapping in each replicate. Specifically, we introduce a Dirichlet prior distribution to capture the common expression pattern of replicates from the same condition, and treat the isoform expression of individual replicates as samples from this distribution. Ambiguous read mapping is modeled as a multinomial distribution, and ambiguous reads are assigned to the most probable isoform in each replicate. Additionally, DEIsoM couples an efficient variational inference and a post-analysis method to improve the accuracy and speed of identification of DE isoforms over alternative methods. Application of DEIsoM to an hepatocellular carcinoma (HCC) dataset identifies biologically relevant DE isoforms. The relevance of these genes/isoforms to HCC are supported by principal component analysis (PCA), read coverage visualization, and the biological literature. AVAILABILITY AND IMPLEMENTATION: The software is available at https://github.com/hao-peng/DEIsoM. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Tensor decomposition methods are effective tools for modelling multidimensional array data (i.e., tensors). Among them, nonparametric Bayesian models, such as Infinite Tucker Decomposition (InfTucker), are more powerful than multilinear factorization approaches, including Tucker and PARAFAC, and usually achieve better predictive performance. However, they are difficult to handle massive data due to a prohibitively high training cost. To address this limitation, we propose Distributed infinite Tucker (DinTucker), a new hierarchical Bayesian model that enables local learning of InfTucker on subarrays and global information integration from local results. We further develop a distributed stochastic gradient descent algorithm, coupled with variational inference for model estimation. In addition, the connection between DinTucker and InfTucker is revealed in terms of model evidence. Experiments demonstrate that DinTucker maintains the predictive accuracy of InfTucker and is scalable on massive data: On multidimensional arrays with billions of elements from two real-world applications, DinTucker achieves significantly higher prediction accuracy with less training time, compared with the state-of-the-art large-scale tensor decomposition method, GigaTensor.

Sparse learning has received tremendous amount of interest in high-dimensional data analysis due to its model interpretability and the low-computational cost. Among the various techniques, adaptive l1-regularization is an effective framework to improve the convergence behaviour of the LASSO, by using varying strength of regularization across different features. In the meantime, the adaptive structure makes it very powerful in modelling grouped sparsity patterns as well, being particularly useful in high-dimensional multi-task problems. However, choosing an appropriate, global regularization weight is still an open problem. In this paper, inspired by the annealing technique in material science, we propose to achieve "annealed sparsity" by designing a dynamic shrinking scheme that simultaneously optimizes the regularization weights and model coefficients in sparse (multi-task) learning. The dynamic structures of our algorithm are twofold. Feature-wise (spatially), the regularization weights are updated interactively with model coefficients, allowing us to improve the global regularization structure. Iteration-wise (temporally), such interaction is coupled with gradually boosted l1-regularization by adjusting an equality norm-constraint, achieving an annealing effect to further improve model selection. This renders interesting shrinking behaviour in the whole solution path. Our method competes favorably with state-of-the-art methods in sparse (multi-task) learning. We also apply it in expression quantitative trait loci analysis (eQTL), which gives useful biological insights in human cancer (melanoma) study.

Tensor factorization is a powerful tool to analyse multi-way data. Recently proposed nonlinear factorization methods, although capable of capturing complex relationships, are computationally quite expensive and may suffer a severe learning bias in case of extreme data sparsity. Therefore, we propose a distributed, flexible nonlinear tensor factorization model, which avoids the expensive computations and structural restrictions of the Kronecker-product in the existing TGP formulations, allowing an arbitrary subset of tensor entries to be selected for training. Meanwhile, we derive a tractable and tight variational evidence lower bound (ELBO) that enables highly decoupled, parallel computations and high-quality inference. Based on the new bound, we develop a distributed, key-value-free inference algorithm in the MapReduce framework, which can fully exploit the memory cache mechanism in fast MapReduce systems such as Spark. Experiments demonstrate the advantages of our method over several state-of-the-art approaches, in terms of both predictive performance and computational efficiency.

The analysis and diagnosis of Alzheimer’s disease (AD) can be based on genetic variations, e.g., single nucleotide polymorphisms (SNPs) and phenotypic traits, e.g., Magnetic Resonance Imaging (MRI) features. We consider two important and related tasks: i) to select genetic and phenotypical markers for AD diagnosis and ii) to identify associations between genetic and phenotypical data. While previous studies treat these two tasks separately, they are tightly coupled because underlying associations between genetic variations and phenotypical features contain the biological basis for a disease. Here we present a new sparse Bayesian approach for joint association study and disease diagnosis. In this approach, common latent features are extracted from different data sources based on sparse projection matrices and used to predict multiple disease severity levels; in return, the disease status can guide the discovery of relationships between data sources. The sparse projection matrices not only reveal interactions between data sources but also select groups of biomarkers related to the disease. Moreover, to take advantage of the linkage disequilibrium (LD) measuring the non-random association of alleles, we incorporate a graph Laplacian type of prior in the model. To learn the model from data, we develop an efficient variational inference algorithm. Analysis on an imaging genetics dataset for the study of Alzheimer’s Disease (AD) indicates that our model identifies biologically meaningful associations between genetic variations and MRI features, and achieves significantly higher accuracy for predicting ordinal AD stages than the competing methods.

Supervised dimensionality reduction has shown great advantages in finding predictive subspaces. Previous methods rarely consider the popular maximum margin principle and are prone to overfitting to usually small training data, especially for those under the maximum likelihood framework. In this paper, we present a posterior-regularized Bayesian approach to combine Principal Component Analysis (PCA) with the max-margin learning. Based on the data augmentation idea for max-margin learning and the probabilistic interpretation of PCA, our method can automatically infer the weight and penalty parameter of max-margin learning machine, while finding the most appropriate PCA subspace simultaneously under the Bayesian framework. We develop a fast mean-field variational inference algorithm to approximate the posterior. Experimental results on various classification tasks show that our method outperforms a number of competitors.

In the analysis and diagnosis of many diseases, such as the Alzheimer's disease (AD), two important and related tasks are usually required: i) selecting genetic and phenotypical markers for diagnosis, and ii) identifying associations between genetic and phenotypical features. While previous studies treat these two tasks separately, they are tightly coupled due to the same underlying biological basis. To harness their potential benefits for each other, we propose a new sparse Bayesian approach to jointly carry out the two important and related tasks. In our approach, we extract common latent features from different data sources by sparse projection matrices and then use the latent features to predict disease severity levels; in return, the disease status can guide the learning of sparse projection matrices, which not only reveal interactions between data sources but also select groups of related biomarkers. In order to boost the learning of sparse projection matrices, we further incorporate graph Laplacian priors encoding the valuable linkage disequilibrium (LD) information. To efficiently estimate the model, we develop a variational inference algorithm. Analysis on an imaging genetics dataset for AD study shows that our model discovers biologically meaningful associations between single nucleotide polymorphisms (SNPs) and magnetic resonance imaging (MRI) features, and achieves significantly higher accuracy for predicting ordinal AD stages than competitive methods.

Multiway data analysis deals with multiway arrays, i.e., tensors, and the goal is twofold: predicting missing entries by modeling the interactions between array elements and discovering hidden patterns, such as clusters or communities in each mode. Despite the success of existing tensor factorization approaches, they are either unable to capture nonlinear interactions, or computationally expensive to handle massive data. In addition, most of the existing methods lack a principled way to discover latent clusters, which is important for better understanding of the data. To address these issues, we propose a scalable nonparametric tensor decomposition model. It employs Dirichlet process mixture (DPM) prior to model the latent clusters; it uses local Gaussian processes (GPs) to capture nonlinear relationships and to improve scalability. An efficient online variational Bayes Expectation-Maximization algorithm is proposed to learn the model. Experiments on both synthetic and real-world data show that the proposed model is able to discover latent clusters with higher prediction accuracy than competitive methods. Furthermore, the proposed model obtains significantly better predictive performance than the state-of-the-art large scale tensor decomposition algorithm, GigaTensor, on two large datasets with billions of entries.

Tensor decomposition is a powerful computational tool for multiway data analysis. Many popular tensor decomposition approaches-such as the Tucker decomposition and CANDECOMP/PARAFAC (CP)-amount to multi-linear factorization. They are insufficient to model (i) complex interactions between data entities, (ii) various data types (e.g., missing data and binary data), and (iii) noisy observations and outliers. To address these issues, we propose tensor-variate latent nonparametric Bayesian models for multiway data analysis. We name these models InfTucker. These new models essentially conduct Tucker decomposition in an infinite feature space. Unlike classical tensor decomposition models, our new approaches handle both continuous and binary data in a probabilistic framework. Unlike previous Bayesian models on matrices and tensors, our models are based on latent Gaussian or t processes with nonlinear covariance functions. Moreover, on network data, our models reduce to nonparametric stochastic blockmodels and can be used to discover latent groups and predict missing interactions. To learn the models efficiently from data, we develop a variational inference technique and explore properties of the Kronecker product for computational efficiency. Compared with a classical variational implementation, this technique reduces both time and space complexities by several orders of magnitude. On real multiway and network data, our new models achieved significantly higher prediction accuracy than state-of-art tensor decomposition methods and blockmodels.

In this paper, we present a nonparametric Bayesian multi-task large-margin classification model which can cluster tasks into the most appropriate number of groups and induce flexible model sharing within each task group simultaneously. Specifically, we first show a very simple method to integrate large margin learning with hierarchical Bayesian models by employing an important variant of the standard SVMi.e.proximal SVM (PSVM)whose loss function is used to define a novel likelihood function. And then we assume that the model parameter of each task consists of two parts: one is shared within each task group (group-level parameter) while the other is specific to each distinct task (task rescaling parameter). A Dirichlet process prior is imposed on the group-level parameter while the task rescaling parameter is assigned a one-mean Laplace prior. Finally the parameter of a task is the corresponding group parameter times its specific rescaling parameter. We give efficient Markov chain Monte Calo (MCMC) algorithm to conduct model inference. Experiments on the Landmine detection data and the UCI Yeast data demonstrate the effectiveness of our method.

One of Android's main defense mechanisms against malicious apps is a risk communication mechanism which, before a user installs an app, warns the user about the permissions the app requires, trusting that the user will make the right decision. This approach has been shown to be ineffective as it presents the risk information of each app in a “stand-alone” fashion and in a way that requires too much technical knowledge and time to distill useful information. We discuss the desired properties of risk signals and relative risk scores for Android apps in order to generate another metric that users can utilize when choosing apps. We present a wide range of techniques to generate both risk signals and risk scores that are based on heuristics as well as principled machine learning techniques. Experimental results conducted using real-world data sets show that these methods can effectively identify malware as very risky, are simple to understand, and easy to use.

MOTIVATION: By capturing various biochemical interactions, biological pathways provide insight into underlying biological processes. Given high-dimensional microarray or RNA-sequencing data, a critical challenge is how to integrate them with rich information from pathway databases to jointly select relevant pathways and genes for phenotype prediction or disease prognosis. Addressing this challenge can help us deepen biological understanding of phenotypes and diseases from a systems perspective. RESULTS: In this article, we propose a novel sparse Bayesian model for joint network and node selection. This model integrates information from networks (e.g. pathways) and nodes (e.g. genes) by a hybrid of conditional and generative components. For the conditional component, we propose a sparse prior based on graph Laplacian matrices, each of which encodes detailed correlation structures between network nodes. For the generative component, we use a spike and slab prior over network nodes. The integration of these two components, coupled with efficient variational inference, enables the selection of networks as well as correlated network nodes in the selected networks. Simulation results demonstrate improved predictive performance and selection accuracy of our method over alternative methods. Based on three expression datasets for cancer study and the KEGG pathway database, we selected relevant genes and pathways, many of which are supported by biological literature. In addition to pathway analysis, our method is expected to have a wide range of applications in selecting relevant groups of correlated high-dimensional biomarkers. AVAILABILITY: The code can be downloaded at www.cs.purdue.edu/homes/szhe/software.html. CONTACT: [email protected].

Online learning has become increasingly popular on handling massive data. The sequential nature of online learning, however, requires a centralized learner to store data and update parameters. In this paper, we consider online learning with distributed data sources. The autonomous learners update local parameters based on local data sources and periodically exchange information with a small subset of neighbors in a communication network. We derive the regret bound for strongly convex functions that generalizes the work by Ram et al. for convex functions. More importantly, we show that our algorithm has intrinsic privacy-preserving properties, and we prove the sufficient and necessary conditions for privacy preservation in the network. These conditions imply that for networks with greater-than-one connectivity, a malicious learner cannot reconstruct the subgradients (and sensitive raw data) of other learners, which makes our algorithm appealing in privacy-sensitive applications.

One of Android's main defense mechanisms against malicious apps is a risk communication mechanism which, before a user installs an app, warns the user about the permissions the app requires, trusting that the user will make the right decision. This approach has been shown to be ineffective as it presents the risk information of each app in a "tand-alone" ashion and in a way that requires too much technical knowledge and time to distill useful information.

Smart electric meters pose a substantial threat to the privacy of individuals in their own homes. Combined with non-intrusive load monitors, smart meter data can reveal precise home appliance usage information. An emerging solution to behavior leakage in smart meter measurement data is the use of battery-based load hiding. In this approach, a battery is used to store and supply power to home devices at strategic times to hide appliance loads from smart meters. A few such battery control algorithms have already been studied in the literature, but none have been evaluated from an adversarial point of view. In this paper, we first consider two well known battery privacy algorithms, Best Effort (BE) and Non-Intrusive Load Leveling (NILL), and demonstrate attacks that recover precise load change information, which can be used to recover appliance behavior information, under both algorithms. We then introduce a stepping approach to battery privacy algorithms that fundamentally differs from previous approaches by maximizing the error between the load demanded by a home and the external load seen by a smart meter. By design, precise load change recovery attacks are impossible. We also propose mutual-information based measurements to evaluate the privacy of different algorithms. We implement and evaluate four novel algorithms using the stepping approach, and show that under the mutual-information metrics they outperform BE and NILL.

We present a new direction for semi-supervised learning where self-adjusting generative models replace fixed ones and unlabeled data can potentially improve learning even when labeled data is only partially-observed. We model each class data by a mixture model and use a hierarchical Dirichlet process (HDP) to model observed as well as unobserved classes. We extend the standard HDP model to accommodate unlabeled samples and introduce a new sharing strategy, within the context of Gaussian mixture models, that restricts sharing with covariance matrices while leaving the mean vectors free. Our research is mainly driven by real-world applications with evolving data-generating mechanisms where obtaining a fully-observed labeled data set is impractical. We demonstrate the feasibility of the proposed approach for semi-supervised learning in two such applications.

We consider the problem of modeling network interactions and identifying latent groups of network nodes. This problem is challenging due to the facts i) that the network nodes are interdependent instead of independent, ii) that the network data are very noisy (e.g., missing edges), and iii) that the network interactions are often sparse. To address these challenges, we propose a Sparse Matrix-variate t process Blockmodel (SMTB). In particular, we generalize a matrix-variate t distribution to a t process on matrices with nonlinear covariance functions. Due to this generalization, our model can estimate latent memberships for individual network nodes. This separates our model from previous t distribution based relational models. Also, we introduce sparse prior distributions on the latent membership parameters to select group assignments for individual nodes. To learn the model efficiently from data, we develop a variational method. When compared with several state-of-the-art models, including the predictive matrix-variate t models and mixed membership stochastic blockmodels, our model achieved improved prediction accuracy on real world network datasets.

Approximate inference is an important technique for dealing with large, intractable graphical models based on the exponential family of distributions. We extend the idea of approximate inference to the t-exponential family by defining a new t-divergence. This divergence measure is obtained via convex duality between the log-partition function of the t-exponential family and a new t-entropy. We illustrate our approach on the Bayes Point Machine with a Student's t-prior.

For many real-world applications, we often need to select correlated variables---such as genetic variations and imaging features associated with Alzheimer's disease---in a high dimensional space. The correlation between variables presents a challenge to classical variable selection methods. To address this challenge, the elastic net has been developed and successfully applied to many applications. Despite its great success, the elastic net does not exploit the correlation information embedded in the data to select correlated variables. To overcome this limitation, we present a novel hybrid model, EigenNet, that uses the eigenstructures of data to guide variable selection. Specifically, it integrates a sparse conditional classification model with a generative model capturing variable correlations in a principled Bayesian framework. We develop an efficient active-set algorithm to estimate the model via evidence maximization. Experiments on synthetic data and imaging genetics data demonstrated the superior predictive performance of the EigenNet over the lasso, the elastic net, and the automatic relevance determination.

There has been increasing interest in automatic techniques for generating roles for role based access control, a process known as role mining. Most role mining approaches assume the input data is clean, and attempt to optimize the RBAC state. We examine role mining with noisy input data and suggest dividing the problem into two steps: noise removal and candidate role generation. We introduce an approach to use (non-binary) rank reduced matrix factorization to identify noise and experimentally show that it is effective at identifying noise in access control data. User- and permission-attributes can further be used to improve accuracy. Next, we show that our two-step approach is able to find candidate roles that are close to the roles mined from noise-less data. This method performs better than the approach of mining noisy data directly and offering the administrator increased control in the noise removal and candidate role generation phases. We note that our approach is applicable outside role engineering and may be used to identify errors or predict missing values in any access control matrix.

The recent emergence of Graphics Processing Units (GPUs) as general-purpose parallel computing devices provides us with new opportunities to develop scalable learning methods for massive data. In this work, we consider the problem of parallelizing two inference methods on GPUs for latent Dirichlet Allocation (LDA) models, collapsed Gibbs sampling (CGS) and collapsed variational Bayesian (CVB). To address limited memory constraints on GPUs, we propose a novel data partitioning scheme that effectively reduces the memory cost. Furthermore, the partitioning scheme balances the computational cost on each multiprocessor and enables us to easily avoid memory access conflicts. We also use data streaming to handle extremely large datasets. Extensive experiments showed that our parallel inference methods consistently produced LDA models with the same predictive power as sequential training methods did but with 26x speedup for CGS and 196x speedup for CVB on a GPU with 30 multiprocessors; actually the speedup is almost linearly scalable with the number of multiprocessors available. The proposed partitioning scheme and data streaming can be easily ported to many other models in machine learning.

In this paper, we propose a new Bayesian receiver for signal detection in flat-fading channels. First, the detection problem is formulated as an inference problem in a graphical model that models a hybrid dynamic system with both continuous and discrete variables. Then, based on the expectation propagation (EP) framework, we develop a smoothing algorithm to address the inference problem and visualize this algorithm using factor graphs. As a generalization of loopy belief propagation, EP efficiently approximates Bayesian estimation by iteratively propagating information between different nodes in the graphical model and projecting the posterior distributions into the exponential family. We use window-based EP smoothing for online estimation as in the signal detection problem. Window-based EP smoothing achieves accuracy similar to that obtained by batch EP smoothing, as shown in our simulations, while reducing delay time. Compared to sequential Monte Carlo filters and smoothers, the new method has lower computational complexity since it makes analytically deterministic approximation instead of Monte Carlo approximations. Our simulations demonstrate that the new receiver achieves accurate detection without the aid of any training symbols or decision feedbacks. Furthermore, the new receiver achieves accuracy comparable to that achieved by sequential Monte Carlo methods, but with less than one-tenth computational cost.

Bayesian inference has become increasingly important in statistical machine learning. Exact Bayesian calculations are often not feasible in practice, however. A number of approximate Bayesian methods have been proposed to make such calculations practical, among them the variational Bayesian (VB) approach. The VB approach, while useful, can nevertheless suffer from slow convergence to the approximate solution. To address this problem, we propose Parameter-eXpanded Variational Bayesian (PX-VB) methods to speed up VB. The new algorithm is inspired by parameter-expanded expectation maximization (PX-EM) and parameterexpanded data augmentation (PX-DA). Similar to PX-EM and -DA, PX-VB expands a model with auxiliary variables to reduce the coupling between variables in the original model. We analyze the convergence rates of VB and PX-VB and demonstrate the superior convergence rates of PX-VB in variational probit regression and automatic relevance determination.

DOAJ is a unique and extensive index of diverse open access journals from around the world, driven by a growing community, committed to ensuring quality content is freely available online for everyone.

Hand-drawn diagrams present a complex recognition problem. Elements of the diagram are often individually ambiguous, and require context to be interpreted. We present a recognition method based on Bayesian conditional random fields (BCRFs) that jointly analyzes all drawing elements in order to incorporate contextual cues. The classification of each object affects the classification of its neighbors. BCRFs allow flexible and correlated features, and take both spatial and temporal information into account. BCRFs estimate the posterior distribution of parameters during training, and average predictions over the posterior for testing. As a result of model averaging, BCRFs avoid the overfitting problems associated with maximum likelihood training. We also incorporate automatic relevance determination (ARD), a Bayesian feature selection technique, into BCRFs. The result is significantly lower error rates compared to ML- and MAP-trained CRFs.

In digital communications, symbol detection in phase noise is an important topic that has been discussed in many papers under different conditions. In this paper, we consider symbol detection with time-varying unknown phase. We propose a solution based on expectation propagation (EP). EP is an extension to belief propagation and developed in machine learning. We point out that the developed EP solution can be considered as an iterated extended Kalman smoother (EKS). However, a crucial step of recycling the likelihoods in EP makes possible the further improvement over EKS. We show in the simulation that EP can produce very good performance with relatively low complexity. Since it produce soft information, the EP solution can be readily applied to iterative detection of coded systems.

There have been many graph-based approaches for semi-supervised clas- sification. One problem is that of hyperparameter learning: performance depends greatly on the hyperparameters of the similarity graph, trans- formation of the graph Laplacian and the noise model. We present a Bayesian framework for learning hyperparameters for graph-based semi- supervised classification. Given some labeled data, which can contain inaccurate labels, we pose the semi-supervised classification as an in- ference problem over the unknown labels. Expectation Propagation is used for approximate inference and the mean of the posterior is used for classification. The hyperparameters are learned using EM for evidence maximization. We also show that the posterior mean can be written in terms of the kernel matrix, providing a Bayesian classifier to classify new points. Tests on synthetic and real datasets show cases where there are significant improvements in performance over the existing approaches.

In many real-world classification problems the input contains a large number of potentially ir-relevant features. This paper proposes a new Bayesian framework for determining the rele-vance of input features. This approach extends one of the most successful Bayesian methods for feature selection and sparse learning, known as Automatic Relevance Determination (ARD). ARD finds the relevance of features by optimiz-ing the model marginal likelihood, also known as the evidence. We show that this can lead to over-fitting. To address this problem, we propose Pre-dictive ARD based on estimating the predictive performance of the classifier. While the actual leave-one-out predictive performance is generally very costly to compute, the expectation propaga-tion (EP) algorithm proposed by Minka provides an estimate of this predictive performance as a side-effect of its iterations. We exploit this in our algorithm to do feature selection, and to select data points in a sparse Bayesian kernel classifier. Moreover, we provide two other improvements to previous algorithms, by replacing Laplace’s approximation with the generally more accurate EP, and by incorporating the fast optimization algorithm proposed by Faul and Tipping. Our experiments show that our method based on the EP estimate of predictive performance is more accurate on test data than relevance determina-tion by optimizing the evidence.

Approximation structure plays an important role in inference on loopy graphs. As a tractable structure, tree approximations have been utilized in the variational method of Ghahramani & Jordan (1997) and the se- quential projection method of Frey et al. (2000). However, belief propa- gation represents each factor of the graph with a product of single-node messages. In this paper, belief propagation is extended to represent fac- tors with tree approximations, by way of the expectation propagation framework. That is, each factor sends a “message” to all pairs of nodes in a tree structure. The result is more accurate inferences and more fre- quent convergence than ordinary belief propagation, at a lower cost than variational trees or double-loop algorithms.

Spectral estimation methods typically assume stationarity and uniform spacing between samples of data. The non-stationarity of real data is usually accommodated by windowing methods, while the lack of uniformly-spaced samples is typically addressed by methods that “fill in” the data in some way. This paper presents a new approach to both of these problems: We use a non-stationary Kalman filter within a Bayesian framework to jointly estimate all spectral coefficients instantaneously. The new method works regardless of how the signal samples are spaced. We illustrate the method on several data sets, showing that it provides more accurate estimation than the Lomb-Scargle method and several classical spectral estimation methods.

We propose a new hybrid unsupervised/supervised learning scheme that integrates independent component analysis (ICA) with the support vector machine (SVM) approach and apply this new learning scheme to the face detection problem. In low-level feature extraction, ICA produces independent image bases that emphasize edge information in the image data. In high-level classification, SVM classifies the ICA features as a face or non-faces. Our experimental results show that by using ICA features we obtain a larger margin of separation and fewer support vectors than by training SVM directly on the image data. This indicates better generalization performance, which is verified in our experiments.