Author name cluster
Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.
EAAI Journal 2026 Journal Article
EAAI Journal 2026 Journal Article
EAAI Journal 2026 Journal Article
JBHI Journal 2026 Journal Article
Knee osteoarthritis (KOA) is a prevalent musculoskeletal disorder, often diagnosed using X-rays due to its cost-effectiveness. While Magnetic Resonance Imaging (MRI) provides superior soft tissue visualization and serves as a valuable supplementary diagnostic tool, its high cost and limited accessibility significantly restrict its widespread use. To explore the feasibility of bridging this imaging gap, we conducted a feasibility study leveraging a diffusion-based model that uses an X-ray image as conditional input, alongside target depth and additional patient-specific feature information, to generate corresponding MRI sequences. Our findings demonstrate that the MRI volumes generated by our approach are not only visually closer to real MRI scans compared with other methods but also achieve the highest quantitative performance in terms of Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM). Furthermore, by increasing the number of inference steps to interpolate between slice depths, we enhance the continuity of the generated volume, achieving higher adjacent slice correlation coefficients. Through ablation studies, we further validate that integrating supplemental patient-specific information, beyond what X-rays alone can provide, enhances the accuracy and clinical relevance of the generated MRI, which underscores the potential of leveraging external patient-specific information to improve the performance of the MRI generation.
AAAI Conference 2026 Conference Paper
All-in-one image restoration (AIR) aims to address diverse degradations within a unified model by leveraging informative degradation conditions to guide the restoration process. However, existing methods often rely on implicitly learned priors, which may entangle feature representations and hinder performance in complex or unseen scenarios. Histogram of Oriented Gradients (HOG) as a classical gradient representation, we observe that it has strong discriminative capability across diverse degradations, making it a powerful and interpretable prior for AIR. Based on this insight, we propose HOGformer, a Transformer-based model that integrates learnable HOG features for degradation-aware restoration. The core of HOGformer is a Dynamic HOG-aware Self-Attention (DHOGSA) mechanism, which adaptively models long-range spatial dependencies conditioned on degradation-specific cues encoded by HOG descriptors. To further adapt the heterogeneity of degradations in AIR, we propose a Dynamic Interaction Feed-Forward (DIFF) module that facilitates channel–spatial interactions, enabling robust feature transformation under diverse degradations. Besides, we propose a HOG loss to explicitly enhance structural fidelity and edge sharpness. Extensive experiments on a variety of benchmarks, including adverse weather and natural degradations, demonstrate that HOGformer achieves state-of-the-art performance and generalizes well to complex real-world scenarios.
EAAI Journal 2026 Journal Article
EAAI Journal 2026 Journal Article
AIIM Journal 2026 Journal Article
AIIM Journal 2025 Journal Article
NeurIPS Conference 2025 Conference Paper
Heterogeneous hypergraph is a kind of structural data that contains multiple types of nodes and multiple types of hyperedges. Each hyperedge type corresponds to a specific multi-ary relation (called hyper-relation) among subsets of nodes, which goes beyond traditional pair-wise relations in simple graphs. Existing representation learning methods for heterogeneous hypergraphs typically learn embeddings for nodes and hyperedges based on graph neural networks. Although achieving promising performance, they are still limited in capturing more complex structural features and richer semantics conveyed by the composition of various hyper-relations. To fill this research gap, in this work, we propose the concept of hyper-meta-path for heterogeneous hypergraphs, which is defined as the composition of a sequence of hyper-relations. Besides, we design an attention-based heterogeneous hypergraph neural network (HHNN) to automatically learn the importance of hyper-meta-paths. By exploiting useful ones, HHNN is able to capture more complex structural features to boost the model's performance, as well as leverage their conveyed semantics to improve the model's interpretability. Extensive experiments show that HHNN can achieve significantly better performance than state-of-the-art baselines, and the discovered hyper-meta-paths bring good interpretability for the model predictions. To facilitate the reproducibility of this work, we provide our dataset as well as anonymized source code at: https: //github. com/zhengziyu77/HHNN.
EAAI Journal 2025 Journal Article
AAAI Conference 2025 Conference Paper
Instruction tuning has emerged as an effective approach that notably improves large language models (LLMs) performance, showing particular promise in natural language generation tasks by producing more diverse, coherent, and task-relevant outputs. However, extending instruction tuning to natural language understanding (NLU) tasks presents significant challenges, primarily due to the difficulty in achieving high-precision responses and the scarcity of large-scale, high-quality instruction data necessary for effective tuning. In this work, we introduce Adversarial Noisy Instruction Tuning (ANIT) to improve NLU performance on LLMs. First, we leverage low-resource techniques to construct noisy instruction datasets. Second, we employ semantic distortion-aware techniques to quantify the intensity of noise within these instructions. Last, we devise an adversarial training method that incorporates a noise response strategy to achieve noisy instruction tuning. ANIT enhances LLMs capability to detect and accommodate semantic distortions in noisy instructions, thereby augmenting their comprehension of task objectives and ability to generate more accurate responses. We evaluate our approach across diverse noisy instructions and semantic distortion quantification methods on multiple NLU tasks. Comprehensive empirical results demonstrate that our method consistently outperforms existing approaches across various experimental settings.
ECAI Conference 2025 Conference Paper
Electroencephalography has been validated as an effective technique for detecting Parkinson’s disease, particularly in its early stages. However, the high cost of EEG data annotation often results in limited dataset size and considerable discrepancies across datasets, including differences in acquisition protocols and subject demographics, significantly hinder the robustness and generalizability of models in cross-dataset detection scenarios. To address such challenges, this paper proposes a semi-supervised learning framework named MCLPD, which integrates multi-view contrastive pre-training with lightweight supervised fine-tuning to enhance cross-dataset PD detection performance. During pre-training, MCLPD uses self-supervised learning on the unlabeled UNM dataset. To build contrastive pairs, it applies dual augmentations in both time and frequency domains, which enrich the data and naturally fuse time-frequency information. In the fine-tuning phase, only a small proportion of labeled data from another two datasets (UI and UC) is used for supervised optimization. Experimental results show that MCLPD achieves F1 scores of 0. 91 on UI and 0. 81 on UC using only 1% of labeled data, which further improve to 0. 97 and 0. 87, respectively, when 5% of labeled data is used. Compared to existing methods, MCLPD substantially improves cross-dataset generalization while reducing the dependency on labeled data, demonstrating the effectiveness of the proposed framework.
AAAI Conference 2025 Conference Paper
Missense mutations could affect the Liquid-Liquid Phase Separation (LLPS) propensity of proteins and lead to aberrant phase-separating behaviours, which are recently found to be associated with many diseases including Alzheimer's and cancer. However, the regulatory role of mutations in LLPS remains unclear due to challenges in accurately characterizing the LLPS ability of mutants, including the high similarity in features, lack of labeled data, and vast amounts of data involved. To bridge this gap and facilitate the discovery of therapeutic strategies, we propose the first machine learning-based guider for protein phase-separating behaviour alteration, PScalpel. PScalpel leverages both structural information and an auxiliary tasks-based graph contrastive learning framework to distinguish the mutants’ LLPS ability, and incorporates a genetic algorithms-based recommendation method to identify mutants with desired LLPS properties. Comprehensive computational and biological experiments validate the effectiveness of PScalpel as a versatile tool for guiding alterations in protein phase separation behavior.
NeurIPS Conference 2025 Conference Paper
We introduce PurpCode, the first post-training recipe for training safe code reasoning models towards generating secure code and defending against malicious cyberactivities. PurpCode trains a reasoning model in two stages: (i) Rule Learning, which explicitly teaches the model to reference cybersafety rules to generate vulnerability-free code and to avoid facilitating malicious cyberactivities; and (ii) Reinforcement Learning, which optimizes model safety and preserves model utility through diverse, multi-objective reward mechanisms. To empower the training pipelines with comprehensive cybersafety data, we conduct internal red-teaming to synthesize comprehensive and high-coverage prompts based on real-world tasks for inducing unsafe cyberactivities in the model. Based on PurpCode, we develop a reasoning-based coding model, namely PurpCode-32B, which demonstrates state-of-the-art cybersafety, outperforming various frontier models. Moreover, our alignment method decreases the model overrefusal rates in both general and cybersafety-specific scenarios, while preserving model utility in both code generation and common security knowledge.
AAAI Conference 2025 Conference Paper
Automatic Radiology Report Generation (RRG) is an important topic for alleviating the substantial workload of radiologists. Existing RRG approaches rely on supervised regression based on different architectures or additional knowledge injection, while the generated report may not align optimally with radiologists’ preferences. Especially, since the preferences of radiologists are inherently heterogeneous and multi-dimensional, e.g., some may prioritize report fluency, while others emphasize clinical accuracy. To address this problem, we propose a new RRG method via Multi-objective Preference Optimization (MPO) to align the pre-trained RRG model with multiple human preferences, which can be formulated by multi-dimensional reward functions and optimized by multi-objective reinforcement learning (RL). Specifically, we use a preference vector to represent the weight of preferences and use it as a condition for the RRG model. Then, a linearly weighed reward is obtained via a dot product between the preference vector and multi-dimensional reward. Next, the RRG model is optimized to align with the preference vector by optimizing such a reward via RL. In the training stage, we randomly sample diverse preference vectors from the preference space and align the model by optimizing the weighted multi-objective rewards, which leads to an optimal policy on the entire preference space. When inference, our model can generate reports aligned with specific preferences without further fine-tuning. Extensive experiments on two public datasets show the proposed method can generate reports that cater to different preferences in a single model and achieve state-of-the-art performance.
ICRA Conference 2025 Conference Paper
End-to-end autonomous driving with vision-only is not only more cost-effective compared to LiDAR-vision fusion but also more reliable than traditional methods. To achieve a economical and robust purely visual autonomous driving system, we propose RenderWorld, a vision-only end-to-end autonomous driving framework, which generates 3D occupancy labels using a self-supervised gaussian-based Img2Occ Module, then encodes the labels by AM-VAE, and uses world model for forecasting and planning. RenderWorld employs Gaussian Splatting to represent 3D scenes and render 2D images greatly improves segmentation accuracy and reduces GPU memory consumption compared with NeRF-based methods. By applying AM-VAE to encode air and non-air separately, RenderWorld achieves more fine-grained scene element representation, leading to state-of-the-art performance in both 4D occupancy forecasting and motion planning from autoregressive world model.
EAAI Journal 2025 Journal Article
AAAI Conference 2025 Conference Paper
The connections between symbolic rules and neural networks have been explored in various directions, including rule mining through neural networks and rule-based explanation for neural networks. These approaches allow symbolic rules to be extracted from neural network models, which offers explainability to the models. However, the plausibility of the extracted rules is rarely analysed. In this paper, we show that the confidence degrees of extracted rules are generally not high, and we propose a new family of Graph Neural Networks that can be trained with the guidance of rules. Hence, the inference of our model simulates the rule reasoning. Moreover, rules with high confidence degrees can be extracted from the trained model that aligns with the inference of the model, which verifies the effectiveness of the rule guidance. Experimental evaluation of knowledge graph reasoning tasks further demonstrates the effectiveness of our model.
AAAI Conference 2025 Conference Paper
Malignant brain tumors have become an aggressive and dangerous disease that leads to death worldwide. Multi-modal MRI data is crucial for accurate brain tumor segmentation, but missing modalities common in clinical practice can severely degrade the segmentation performance. While incomplete multi-modal learning methods attempt to address this, learning robust and discriminative features from arbitrary missing modalities remains challenging. To address this challenge, we propose a novel Semantic-guided Masked Mutual Learning (SMML) approach to distill robust and discriminative knowledge across diverse missing modality scenarios. Specifically, we propose a novel dual-branch masked mutual learning scheme guided by Hierarchical Consistency Constraints (HCC) to ensure multi-level consistency, thereby enhancing mutual learning in incomplete multi-modal scenarios. The HCC framework comprises a pixel-level constraint that selects and exchanges reliable knowledge to guide the mutual learning process. Additionally, it includes a feature-level constraint that uncovers robust inter-sample and inter-class relational knowledge within the latent feature space. To further enhance multi-modal learning from missing modality data, we integrate a refinement network into each student branch. This network leverages semantic priors from the Segment Anything Model (SAM) to provide supplementary information, effectively complementing the masked mutual learning strategy in capturing auxiliary discriminative knowledge. Extensive experiments on three challenging brain tumor segmentation datasets demonstrate that our method significantly improves performance over state-of-the-art methods in diverse missing modality settings.
EAAI Journal 2025 Journal Article
EAAI Journal 2025 Journal Article
JBHI Journal 2025 Journal Article
Detecting early brain anomalies is crucial for patient prognosis and recovery, but obtaining expert-annotated data is challenging, especially for clinically silent early brain anomalies. Unsupervised brain anomaly detection, which identifies anomalous regions by modeling normal brain patterns, has gained interest for its label efficiency. However, the inherent variability in normal brains and subtle anomalies that closely resemble normal tissue pose challenges for traditional autoencoders in distinguishing anomalies. Denoising AutoEncoder (DAE) methods have been explored to enhance the model's ability, while their success hinges on effective noise generation strategies. In this paper, we introduce a novel, structure-preserving noise generation scheme based on cross-modal CutMix, aiming to enhance the diversity of noise patterns while preserving the anatomical structure of the brain. To enhance the robustness of DAE learning, we propose an ensemble approach featuring dual experts, each incorporating distinct scale of noise. This dual-expert scheme effectively amplifies reconstruction errors in anomalous regions and suppresses false alarms in healthy areas. Additionally, we propose an anatomically-aware bidirectional consistency loss to ensure high-fidelity reconstruction at the regional level, using superpixels for anatomy perception and bidirectional distillation for reliable knowledge transfer. Extensive experiments across two different settings demonstrate the effectiveness and generalization ability of our proposed method.
IJCAI Conference 2025 Conference Paper
Distinguishing key features in complex visual tasks is challenging. A novel approach treats image patches (tokens) as waves. By using both phase and amplitude, it captures richer semantics and specific invariances compared to pixel-based methods, and allows for feature fusion across regions for a holistic image representation. Based on this, we propose the Wave-wise Discriminative Transformer Tracker (WDT). During tracking, WDT represents features via phase-amplitude separation, enhancement, and mixture. First, we designed a Mutual Exclusive Phase-Amplitude Extractor (MEPAE) to separate phase and amplitude features with distinct semantics, representing spatial target info and background brightness respectively. Then, Wave-wise Feature Augmentation is carried out with two submodules: Phase-Amplitude Feature Augmentation and Mixture. The augmentation module disrupts the separated features in the same batch, and the mixture module recombines them to generate positive and negative waves. The original features are aggregated into the original wave. Positive waves have the same phase but different amplitudes, and negative waves have different phase components. Finally, self-supervised and tracking-supervised losses guide the global and local representation learning for original, positive, and negative waves, enhancing wave-level discrimination. Experiments on five benchmarks prove the effectiveness of our method.
IJCAI Conference 2024 Conference Paper
Medical Image Segmentation (MIS) plays a crucial role in medical therapy planning and robot navigation. Prototype learning methods in MIS focus on generating segmentation masks through pixel-to-prototype comparison. However, current approaches often overlook sample diversity by using a fixed prototype per semantic class and neglect intra-class variation within each input. In this paper, we propose to generate instance-adaptive prototypes for MIS, which integrates a common prototype proposal (CPP) capturing common visual patterns and an instance-specific prototype proposal (IPP) tailored to each input. To further account for the intra-class variation, we propose to guide the IPP generation by re-weighting the intermediate feature map according to their confidence scores. These confidence scores are hierarchically generated using a transformer decoder. Additionally we introduce a novel self-supervised filtering strategy to prioritize the foreground pixels during the training of the transformer decoder. Extensive experiments demonstrate favorable performance of our method.
AAAI Conference 2024 Conference Paper
Click-through rate (CTR) prediction is a vital task in industrial recommendation systems. Most existing methods focus on the network architecture design of the CTR model for better accuracy and suffer from the data sparsity problem. Especially in industrial recommendation systems, the widely applied negative sample down-sampling technique due to resource limitation worsens the problem, resulting in a decline in performance. In this paper, we propose Auxiliary Match Tasks for enhancing Click-Through Rate (AT4CTR) prediction accuracy by alleviating the data sparsity problem. Specifically, we design two match tasks inspired by collaborative filtering to enhance the relevance modeling between user and item. As the "click" action is a strong signal which indicates the user's preference towards the item directly, we make the first match task aim at pulling closer the representation between the user and the item regarding the positive samples. Since the user's past click behaviors can also be treated as the user him/herself, we apply the next item prediction as the second match task. For both the match tasks, we choose the InfoNCE as their loss function. The two match tasks can provide meaningful training signals to speed up the model's convergence and alleviate the data sparsity. We conduct extensive experiments on one public dataset and one large-scale industrial recommendation dataset. The result demonstrates the effectiveness of the proposed auxiliary match tasks. AT4CTR has been deployed in the real industrial advertising system and has gained remarkable revenue.
AAAI Conference 2024 Conference Paper
Few-shot Visual Question Answering (VQA) realizes few-shot cross-modal learning, which is an emerging and challenging task in computer vision. Currently, most of the few-shot VQA methods are confined to simply extending few-shot classification methods to cross-modal tasks while ignoring the spatial distribution properties of multimodal features and cross-modal information interaction. To address this problem, we propose a novel Cross-modal feature Distribution Calibration Inference Network (CDCIN) in this paper, where a new concept named visual information entropy is proposed to realize multimodal features distribution calibration by cross-modal information interaction for more effective few-shot VQA. Visual information entropy is a statistical variable that represents the spatial distribution of visual features guided by the question, which is aligned before and after the reasoning process to mitigate redundant information and improve multi-modal features by our proposed visual information entropy calibration module. To further enhance the inference ability of cross-modal features, we additionally propose a novel pre-training method, where the reasoning sub-network of CDCIN is pretrained on the base class in a VQA classification paradigm and fine-tuned on the few-shot VQA datasets. Extensive experiments demonstrate that our proposed CDCIN achieves excellent performance on few-shot VQA and outperforms state-of-the-art methods on three widely used benchmark datasets.
EAAI Journal 2024 Journal Article
EAAI Journal 2024 Journal Article
ICML Conference 2024 Conference Paper
We introduce Magicoder, a series of fully open-source (code, weights, and data) Large Language Models (LLMs) for code that significantly closes the gap with top code models while having no more than 7B parameters. Magicoder models are trained on 75K synthetic instruction data using OSS-Instruct, a novel approach to enlightening LLMs with open-source code snippets to generate diverse instruction data for code. Our main motivation is to mitigate the inherent bias of the synthetic data generated by LLMs through the wealth of open-source references for the production of more realistic and controllable data. The orthogonality of OSS-Instruct and other data generation methods like Evol-Instruct further enables us to build an enhanced MagicoderS. Both Magicoder and MagicoderS substantially outperform state-of-the-art code models with similar or even larger sizes on a wide range of coding benchmarks. Notably, MagicoderS-CL-7B based on CodeLlama even surpasses the prominent ChatGPT on HumanEval+ (66. 5 vs. 65. 9 in pass@1 ). Overall, OSS-Instruct opens a new direction for crafting diverse synthetic instruction data for code using abundant open-source references.
AAMAS Conference 2024 Conference Paper
DeGroot-style opinion formation presumes a continuous interaction among agents of a social network. Hence, it cannot handle agents external to the social network that interact only temporarily with the permanent ones. Many real-world organisations and individuals fall into such a category. For instance, a company tries to persuade as many as possible to buy its products and, due to various constraints, can only exert its influence for a limited amount of time. We propose a variant of the DeGroot model that allows an external agent to interact with the permanent ones for a preset period of time. We obtain several insights on maximising an external agent’s influence in opinion formation by analysing and simulating the variant.
AAAI Conference 2024 Conference Paper
Class-agnostic motion prediction methods aim to comprehend motion within open-world scenarios, holding significance for autonomous driving systems. However, training a high-performance model in a fully-supervised manner always requires substantial amounts of manually annotated data, which can be both expensive and time-consuming to obtain. To address this challenge, our study explores the potential of semi-supervised learning (SSL) for class-agnostic motion prediction. Our SSL framework adopts a consistency-based self-training paradigm, enabling the model to learn from unlabeled data by generating pseudo labels through test-time inference. To improve the quality of pseudo labels, we propose a novel motion selection and re-generation module. This module effectively selects reliable pseudo labels and re-generates unreliable ones. Furthermore, we propose two data augmentation strategies: temporal sampling and BEVMix. These strategies facilitate consistency regularization in SSL. Experiments conducted on nuScenes demonstrate that our SSL method can surpass the self-supervised approach by a large margin by utilizing only a tiny fraction of labeled data. Furthermore, our method exhibits comparable performance to weakly and some fully supervised methods. These results highlight the ability of our method to strike a favorable balance between annotation costs and performance. Code will be available at https://github.com/kwwcv/SSMP.
NeurIPS Conference 2024 Conference Paper
Masked (or absorbing) diffusion is actively explored as an alternative to autoregressive models for generative modeling of discrete data. However, existing work in this area has been hindered by unnecessarily complex model formulations and unclear relationships between different perspectives, leading to suboptimal parameterization, training objectives, and ad hoc adjustments to counteract these issues. In this work, we aim to provide a simple and general framework that unlocks the full potential of masked diffusion models. We show that the continuous-time variational objective of masked diffusion models is a simple weighted integral of cross-entropy losses. Our framework also enables training generalized masked diffusion models with state-dependent masking schedules. When evaluated by perplexity, our models trained on OpenWebText surpass prior diffusion language models at GPT-2 scale and demonstrate superior performance on 4 out of 5 zero-shot language modeling tasks. Furthermore, our models vastly outperform previous discrete diffusion models on pixel-level image modeling, achieving 2. 75 (CIFAR-10) and 3. 40 (ImageNet 64x64) bits per dimension that are better than autoregressive models of similar sizes.
NeurIPS Conference 2024 Conference Paper
Current methods for training Binarized Neural Networks (BNNs) heavily rely on the heuristic straight-through estimator (STE), which crucially enables the application of SGD-based optimizers to the combinatorial training problem. Although the STE heuristics and their variants have led to significant improvements in BNN performance, their theoretical underpinnings remain unclear and relatively understudied. In this paper, we propose a theoretically motivated optimization framework for BNN training based on Gaussian variational inference. In its simplest form, our approach yields a non-convex linear programming formulation whose variables and associated gradients motivate the use of latent weights and STE gradients. More importantly, our framework allows us to formulate semidefinite programming (SDP) relaxations to the BNN training task. Such formulations are able to explicitly models pairwise correlations between weights during training, leading to a more accurate optimization characterization of the training problem. As the size of such formulations grows quadratically in the number of weights, quickly becoming intractable for large networks, we apply the Burer-Monteiro approach and only optimize over linear-size low-rank SDP solutions. Our empirical evaluation on CIFAR-10, CIFAR-100, Tiny-ImageNet and ImageNet datasets shows our method consistently outperforming all state-of-the-art algorithms for training BNNs.
TMLR Journal 2023 Journal Article
Gradient descent-ascent (GDA) is a widely used algorithm for minimax optimization. However, GDA has been proved to converge to stationary points for nonconvex minimax optimization, which are suboptimal compared with local minimax points. In this work, we develop cubic regularization (CR) type algorithms that globally converge to local minimax points in nonconvex-strongly-concave minimax optimization. We first show that local minimax points are equivalent to second-order stationary points of a certain envelope function. Then, inspired by the classic cubic regularization algorithm, we propose an algorithm named Cubic-LocalMinimax for finding local minimax points, and provide a comprehensive convergence analysis by leveraging its intrinsic potential function. Specifically, we establish the global convergence of Cubic-LocalMinimax to a local minimax point at a sublinear convergence rate and characterize its iteration complexity. Also, we propose a GDA-based solver for solving the cubic subproblem involved in Cubic-LocalMinimax up to certain pre-defined accuracy, and analyze the overall gradient and Hessian-vector product computation complexities of such an inexact Cubic-LocalMinimax algorithm. Moreover, we propose a stochastic variant of Cubic-LocalMinimax for large-scale minimax optimization, and characterize its sample complexity under stochastic sub-sampling. Experimental results demonstrate faster or comparable convergence speed of our stochastic Cubic-LocalMinimax than the state-of-the-art algorithms such as GDA and Minimax Cubic-Newton. In particular, our stochastic Cubic-LocalMinimax was also faster as compared to several other algorithms for minimax optimization on a particular adversarial loss for training a convolutional neural network on MNIST.
EAAI Journal 2023 Journal Article
EAAI Journal 2023 Journal Article
YNIMG Journal 2023 Journal Article
AAAI Conference 2023 Conference Paper
Recent NLP literature has seen growing interest in improving model interpretability. Along this direction, we propose a trainable neural network layer that learns a global interaction graph between words and then selects more informative words using the learned word interactions. Our layer, we call WIGRAPH, can plug into any neural network-based NLP text classifiers right after its word embedding layer. Across multiple SOTA NLP models and various NLP datasets, we demonstrate that adding the WIGRAPH layer substantially improves NLP models' interpretability and enhances models' prediction performance at the same time.
AAMAS Conference 2023 Conference Paper
Inquiry into the price of anarchy (POA) for auctions is almost confined within the risk-neutral setting. Nonetheless, empirical and experimental studies suggest that real-world agents are more or less risk-averse rather than strictly risk-neutral. In this paper, we study the POA of first-price single-item auctions (FPA) with risk-averse bidders. For completeness, we consider both risk-averse and risk-neutral sellers. In the former, we establish that the POA is 1/2 for both the symmetric FPA and FPA in general. In the latter, we show that the POA can be arbitrarily bad for the symmetric FPA and characterise the conditions for the POA to be constant. In response to a fairness issue in the case of risk-neutral sellers, we propose the notion of suboptimal social welfare. We subsequently derive POA bounds with respect to this new notion where the bounds are parameterised by two variables that capture the value range of the utility function.
IJCAI Conference 2022 Conference Paper
In few-shot learning, methods are enslaved to the scarce labeled data, resulting in suboptimal embedding. Recent studies learn the embedding network by other large-scale labeled data. However, the trained network may give rise to the distorted embedding of target data. We argue two respects are required for an unprecedented and promising solution. We call them Better Embedding and More Shots (BEMS). Suppose we propose to extract embedding from the embedding network. BE maximizes the extraction of general representation and prevents over-fitting information. For this purpose, we introduce the topological relation for global reconstruction, avoiding excessive memorizing. MS maximizes the relevance between the reconstructed embedding and the target class space. In this respect, increasing the number of shots is a pivotal but intractable strategy. As a creative method, we derive the bound of information-theory-based loss function and implicitly achieve infinite shots with negligible cost. A substantial experimental analysis is carried out to demonstrate the state-of-the-art performance. Compared to the baseline, our method improves by up to 10%+. We also prove that BEMS is suitable for both standard pre-trained and meta-learning embedded networks.
NeurIPS Conference 2022 Conference Paper
Training GANs under limited data often leads to discriminator overfitting and memorization issues, causing divergent training. Existing approaches mitigate the overfitting by employing data augmentations, model regularization, or attention mechanisms. However, they ignore the frequency bias of GANs and take poor consideration towards frequency information, especially high-frequency signals that contain rich details. To fully utilize the frequency information of limited data, this paper proposes FreGAN, which raises the model's frequency awareness and draws more attention to synthesising high-frequency signals, facilitating high-quality generation. In addition to exploiting both real and generated images' frequency information, we also involve the frequency signals of real images as a self-supervised constraint, which alleviates the GAN disequilibrium and encourages the generator to synthesis adequate rather than arbitrary frequency signals. Extensive results demonstrate the superiority and effectiveness of our FreGAN in ameliorating generation quality in the low-data regime (especially when training data is less than 100). Besides, FreGAN can be seamlessly applied to existing regularization and attention mechanism models to further boost the performance.
IS Journal 2022 Journal Article
Federated learning is a popular framework that guarantees privacy security. Thereinto, heterogeneity is a challenging barrier. We propose a meta-based federated method with mutually constrained propagation (2MFed) method to cope with this. First, 2MFed gives each client the right to choose the most appropriate model from server-side parallel models as its local model, which tolerates hardware heterogeneity. Second, the server mutually exchanges the information of parallel models under the consistency and disparity constraints to improve robustness, thus eliminating data heterogeneity. Third, we use the power of the meta-based label propagation algorithm to treat the federated training as the federated few-shot problem, which removes the model heterogeneity. Finally, extensive experiments confirm the effectiveness of the proposed method. In addition, we evaluate the proposed method on few-shot tasks and demonstrate its excellent performance further.
KR Conference 2022 Conference Paper
Rule learning from large datasets has regained extensive interest as rules are useful for developing explainable approaches to many applications in knowledge graphs. However, existing methods for rule learning are still limited in terms of scalability and rule quality. This paper presents a new method for learning typed rules by employing entity class information. Our experimental evaluation shows the superiority of our system compared to state-of-the-art rule learners. In particular, we demonstrate the usefulness of typed rules in reasoning for link prediction.
NeurIPS Conference 2022 Conference Paper
Recent self-supervised pre-training methods on Heterogeneous Information Networks (HINs) have shown promising competitiveness over traditional semi-supervised Heterogeneous Graph Neural Networks (HGNNs). Unfortunately, their performance heavily depends on careful customization of various strategies for generating high-quality positive examples and negative examples, which notably limits their flexibility and generalization ability. In this work, we present SHGP, a novel Self-supervised Heterogeneous Graph Pre-training approach, which does not need to generate any positive examples or negative examples. It consists of two modules that share the same attention-aggregation scheme. In each iteration, the Att-LPA module produces pseudo-labels through structural clustering, which serve as the self-supervision signals to guide the Att-HGNN module to learn object embeddings and attention coefficients. The two modules can effectively utilize and enhance each other, promoting the model to learn discriminative embeddings. Extensive experiments on four real-world datasets demonstrate the superior effectiveness of SHGP against state-of-the-art unsupervised baselines and even semi-supervised baselines. We release our source code at: https: //github. com/kepsail/SHGP.
NeurIPS Conference 2022 Conference Paper
Despite substantial progress in 3D object detection, advanced 3D detectors often suffer from heavy computation overheads. To this end, we explore the potential of knowledge distillation (KD) for developing efficient 3D object detectors, focusing on popular pillar- and voxel-based detectors. In the absence of well-developed teacher-student pairs, we first study how to obtain student models with good trade offs between accuracy and efficiency from the perspectives of model compression and input resolution reduction. Then, we build a benchmark to assess existing KD methods developed in the 2D domain for 3D object detection upon six well-constructed teacher-student pairs. Further, we propose an improved KD pipeline incorporating an enhanced logit KD method that performs KD on only a few pivotal positions determined by teacher classification response and a teacher-guided student model initialization to facilitate transferring teacher model's feature extraction ability to students through weight inheritance. Finally, we conduct extensive experiments on the Waymo dataset. Our best performing model achieves $65. 75\%$ LEVEL 2 mAPH surpassing its teacher model and requiring only $44\%$ of teacher flops. Our most efficient model runs 51 FPS on an NVIDIA A100, which is $2. 2\times$ faster than PointPillar with even higher accuracy. Code will be available.
AAAI Conference 2021 Conference Paper
Stochastic gradient descent (SGD) is a widely used method for its outstanding generalization ability and simplicity. Adaptive gradient methods have been proposed to further accelerate the optimization process. In this paper, we revisit existing adaptive gradient optimization methods with a new interpretation. Such new perspective leads to a refreshed understanding of the roles of second moments in stochastic optimization. Based on this, we propose Angle-Calibration Moment method (ACMo), a novel stochastic optimization method. It enjoys the benefits of second moments with only first moment updates. Theoretical analysis shows that ACMo is able to achieve the same convergence rate as mainstream adaptive methods. Experiments on a variety of CV and NLP tasks demonstrate that ACMo has a comparable convergence to state-of-the-art Adam-type optimizers, and even a better generalization performance in most cases. The code is available at https: //github. com/Xunpeng746/ACMo.
AAAI Conference 2021 Conference Paper
Recently proposed knowledge distillation approaches based on feature-map transfer validate that intermediate layers of a teacher model can serve as effective targets for training a student model to obtain better generalization ability. Existing studies mainly focus on particular representation forms for knowledge transfer between manually specified pairs of teacher-student intermediate layers. However, semantics of intermediate layers may vary in different networks and manual association of layers might lead to negative regularization caused by semantic mismatch between certain teacherstudent layer pairs. To address this problem, we propose Semantic Calibration for Cross-layer Knowledge Distillation (SemCKD), which automatically assigns proper target layers of the teacher model for each student layer with an attention mechanism. With a learned attention distribution, each student layer distills knowledge contained in multiple layers rather than a single fixed intermediate layer from the teacher model for appropriate cross-layer supervision in training. Consistent improvements over state-of-the-art approaches are observed in extensive experiments with various network architectures for teacher and student models, demonstrating the effectiveness and flexibility of the proposed attention based soft layer association mechanism for cross-layer distillation.
JBHI Journal 2021 Journal Article
Based on the current research on EEG emotion recognition, there are some limitations, such as hand-engineered features, redundant and meaningless signal frames and the loss of frame-to-frame correlation. In this paper, a novel deep learning framework is proposed, named the frame-level distilling neural network (FLDNet), for learning distilled features from the correlations of different frames. A layer named the frame gate is designed to integrate weighted semantic information on multiple frames to remove redundant and meaningless signal frames. A triple-net structure is introduced to distill the learned features net by net to replace the hand-engineered features with professional knowledge. Specifically, one neural network is normally trained for several epochs. Then, a second network of the same structure will be initialized again to learn the extracted features from the frame gate of the first neural network based on the output of the first net. Similarly, the third net improves the features based on the frame gate of the second network. To utilize the representation ability of the triple neural network, an ensemble layer is conducted to integrate the discriminative ability of the proposed framework for final decisions. Consequently, the proposed FLDNet provides an effective method for capturing the correlation between different frames and automatically learn distilled high-level features for emotion recognition. The experiments are carried out in a subject-independent emotion recognition task on public emotion datasets of DEAP and DREAMER benchmarks, which have demonstrated the effectiveness and robustness of the proposed FLDNet.
JAIR Journal 2021 Journal Article
The rapid progress in artificial intelligence (AI) and machine learning has opened unprecedented analytics possibilities in various team and individual sports, including baseball, basketball, and tennis. More recently, AI techniques have been applied to football, due to a huge increase in data collection by professional teams, increased computational power, and advances in machine learning, with the goal of better addressing new scientific challenges involved in the analysis of both individual players’ and coordinated teams’ behaviors. The research challenges associated with predictive and prescriptive football analytics require new developments and progress at the intersection of statistical learning, game theory, and computer vision. In this paper, we provide an overarching perspective highlighting how the combination of these fields, in particular, forms a unique microcosm for AI research, while offering mutual benefits for professional teams, spectators, and broadcasters in the years to come. We illustrate that this duality makes football analytics a game changer of tremendous value, in terms of not only changing the game of football itself, but also in terms of what this domain can mean for the field of AI. We review the state-of-the-art and exemplify the types of analysis enabled by combining the aforementioned fields, including illustrative examples of counterfactual analysis using predictive models, and the combination of game-theoretic analysis of penalty kicks with statistical learning of player attributes. We conclude by highlighting envisioned downstream impacts, including possibilities for extensions to other sports (real and virtual).
EAAI Journal 2021 Journal Article
AAAI Conference 2021 Conference Paper
The end-to-end Human Mesh Recovery (HMR) approach (Kanazawa et al. 2018) has been successfully used for 3D body reconstruction. However, most HMR-based frameworks reconstruct human body by directly learning mesh parameters from images or videos, while lacking explicit guidance of 3D human pose in visual data. As a result, the generated mesh often exhibits incorrect pose for complex activities. To tackle this problem, we propose to exploit 3D pose to calibrate human mesh. Specifically, we develop two novel Pose Calibration frameworks, i. e. , Serial PC-HMR and Parallel PC-HMR. By coupling advanced 3D pose estimators and HMR in a serial or parallel manner, these two frameworks can effectively correct human mesh with guidance of a concise pose calibration module. Furthermore, since the calibration module is designed via non-rigid pose transformation, our PC- HMR frameworks can flexibly tackle bone length variations to alleviate misplacement in the calibrated mesh. Finally, our frameworks are based on generic and complementary integration of data-driven learning and geometrical modeling. Via plug-and-play modules, they can be efficiently adapted for both image/video-based human mesh recovery. Additionally, they have no requirement of extra 3D pose annotations in the testing phase, which releases inference difficulties in practice. We perform extensive experiments on the popular benchmarks, i. e. , Human3. 6M, 3DPW and SURREAL, where our PC-HMR frameworks achieve the SOTA results.
EAAI Journal 2021 Journal Article
AAAI Conference 2020 Conference Paper
A major challenge for video semantic segmentation is the lack of labeled data. In most benchmark datasets, only one frame of a video clip is annotated, which makes most supervised methods fail to utilize information from the rest of the frames. To exploit the spatio-temporal information in videos, many previous works use pre-computed optical flows, which encode the temporal consistency to improve the video segmentation. However, the video segmentation and optical flow estimation are still considered as two separate tasks. In this paper, we propose a novel framework for joint video semantic segmentation and optical flow estimation. Semantic segmentation brings semantic information to handle occlusion for more robust optical flow estimation, while the nonoccluded optical flow provides accurate pixel-level temporal correspondences to guarantee the temporal consistency of the segmentation. Moreover, our framework is able to utilize both labeled and unlabeled frames in the video through joint training, while no additional calculation is required in inference. Extensive experiments show that the proposed model makes the video semantic segmentation and optical flow estimation benefit from each other and outperforms existing methods under the same settings in both tasks.
NeurIPS Conference 2020 Conference Paper
The actor-critic (AC) algorithm is a popular method to find an optimal policy in reinforcement learning. In the infinite horizon scenario, the finite-sample convergence rate for the AC and natural actor-critic (NAC) algorithms has been established recently, but under independent and identically distributed (i. i. d. ) sampling and single-sample update at each iteration. In contrast, this paper characterizes the convergence rate and sample complexity of AC and NAC under Markovian sampling, with mini-batch data for each iteration, and with actor having general policy class approximation. We show that the overall sample complexity for a mini-batch AC to attain an $\epsilon$-accurate stationary point improves the best known sample complexity of AC by an order of $\mathcal{O}(\epsilon^{-1}\log(1/\epsilon))$, and the overall sample complexity for a mini-batch NAC to attain an $\epsilon$-accurate globally optimal point improves the existing sample complexity of NAC by an order of $\mathcal{O}(\epsilon^{-2}/\log(1/\epsilon))$. Moreover, the sample complexity of AC and NAC characterized in this work outperforms that of policy gradient (PG) and natural policy gradient (NPG) by a factor of $\mathcal{O}((1-\gamma)^{-3})$ and $\mathcal{O}((1-\gamma)^{-4}\epsilon^{-2}/\log(1/\epsilon))$, respectively. This is the first theoretical study establishing that AC and NAC attain orderwise performance improvement over PG and NPG under infinite horizon due to the incorporation of critic.
EAAI Journal 2020 Journal Article
EAAI Journal 2020 Journal Article
IJCAI Conference 2020 Conference Paper
Various types of parameter restart schemes have been proposed for proximal gradient algorithm with momentum to facilitate their convergence in convex optimization. However, under parameter restart, the convergence of proximal gradient algorithm with momentum remains obscure in nonconvex optimization. In this paper, we propose a novel proximal gradient algorithm with momentum and parameter restart for solving nonconvex and nonsmooth problems. Our algorithm is designed to 1) allow for adopting flexible parameter restart schemes that cover many existing ones; 2) have a global sub-linear convergence rate in nonconvex and nonsmooth optimization; and 3) have guaranteed convergence to a critical point and have various types of asymptotic convergence rates depending on the parameterization of local geometry in nonconvex and nonsmooth optimization. Numerical experiments demonstrate the convergence and effectiveness of our proposed algorithm.
IJCAI Conference 2020 Conference Paper
Existential rules are an expressive ontology formalism for ontology-mediated query answering and thus query answering is of high complexity, while several tractable fragments have been identified. Existing systems based on first-order rewriting methods can lead to queries too large for DBMS to handle. It is shown that datalog rewriting can result in more compact queries, yet previously proposed datalog rewriting methods are mostly inefficient for implementation. In this paper, we fill the gap by proposing an efficient datalog rewriting approach for answering conjunctive queries over existential rules, and identify and combine existing fragments of existential rules for which our rewriting method terminates. We implemented a prototype system Drewer, and experiments show that it is able to handle a wide range of benchmarks in the literature. Moreover, Drewer shows superior or comparable performance over state-of-the-art systems on both the compactness of rewriting and the efficiency of query answering.
AAAI Conference 2020 Conference Paper
In this paper, we study the problem of query answering with guarded existential rules (also called GNTGDs) under stable model semantics. Our goal is to use existing answer set programming (ASP) solvers. However, ASP solvers handle only finitely-ground logic programs while the program translated from GNTGDs by Skolemization is not in general. To address this challenge, we introduce two novel notions of (1) guarded instantiation forest to describe the instantiation of GNTGDs and (2) prime block to characterize the repeated infinitely-ground program translated from GNTGDs. Using these notions, we prove that the ground termination problem for GNTGDs is decidable. We also devise an algorithm for query answering with GNTGDs using ASP solvers. We have implemented our approach in a prototype system. The evaluation over a set of benchmarks shows encouraging results.
JMLR Journal 2020 Journal Article
Community detection in large social networks is affected by degree heterogeneity of nodes. The D-SCORE algorithm for directed networks was introduced to reduce this effect by taking the element-wise ratios of the singular vectors of the adjacency matrix before clustering. Meaningful results were obtained for the statistician citation network, but rigorous analysis on its performance was missing. First, this paper establishes theoretical guarantee for this algorithm and its variants for the directed degree-corrected block model (Directed-DCBM). Second, this paper provides significant improvements for the original D-SCORE algorithms by attaching the nodes outside of the community cores using the information of the original network instead of the singular vectors. [abs] [ pdf ][ bib ] © JMLR 2020. ( edit, beta )
JAIR Journal 2019 Journal Article
AGM contraction and revision assume an underlying logic that contains propositional logic. Consequently, this assumption excludes many useful logics such as the Horn fragment of propositional logic and most description logics. Our goal in this paper is to generalise AGM contraction and revision to (near-)arbitrary fragments of classical first-order logic. To this end, we first define a very general logic that captures these fragments. In so doing, we make the modest assumptions that a logic contains conjunction and that information is expressed by closed formulas or sentences. The resulting logic is called first-order conjunctive logic or FC logic for short. We then take as the point of departure the AGM approach of constructing contraction functions through epistemic entrenchment, that is the entrenchment-based contraction. We redefine entrenchment-based contraction in ways that apply to any FC logic, which we call FC contraction. We prove a representation theorem showing its compliance with all the AGM contraction postulates except for the controversial recovery postulate. We also give methods for constructing revision functions through epistemic entrenchment which we call FC revision; which also apply to any FC logic. We show that if the underlying FC logic contains tautologies then FC revision complies with all the AGM revision postulates. Finally, in the context of FC logic, we provide three methods for generating revision functions via a variant of the Levi Identity, which we call contraction, withdrawal and cut generated revision, and explore the notion of revision equivalence. We show that withdrawal and cut generated revision coincide with FC revision and so does contraction generated revision under a finiteness condition.
AAAI Conference 2019 Conference Paper
Constructing of molecular structural models from Cryo- Electron Microscopy (Cryo-EM) density volumes is the critical last step of structure determination by Cryo-EM technologies. Methods have evolved from manual construction by structural biologists to perform 6D translation-rotation searching, which is extremely compute-intensive. In this paper, we propose a learning-based method and formulate this problem as a vision-inspired 3D detection and pose estimation task. We develop a deep learning framework for amino acid determination in a 3D Cryo-EM density volume. We also design a sequence-guided Monte Carlo Tree Search (MCTS) to thread over the candidate amino acids to form the molecular structure. This framework achieves 91% coverage on our newly proposed dataset and takes only a few minutes for a typical structure with a thousand amino acids. Our method is hundreds of times faster and several times more accurate than existing automated solutions without any human intervention.
AAAI Conference 2019 Conference Paper
Modal logics are primary formalisms for multi-agent systems but major reasoning tasks in such logics are intractable, which impedes applications of multi-agent modal logics such as automatic planning. One technique of tackling the intractability is to identify a fragment called a normal form of multiagent logics such that it is expressive but tractable for reasoning tasks such as entailment checking, bounded conjunction transformation and forgetting. For instance, DNF of propositional logic is tractable for these reasoning tasks. In this paper, we first introduce a notion of logical separability and then define a novel disjunctive normal form SDNF for the multiagent logic Kn, which overcomes some shortcomings of existing approaches. In particular, we show that every modal formula in Kn can be equivalently casted as a formula in SDNF, major reasoning tasks tractable in propositional DNF are also tractable in SDNF, and moreover, formulas in SDNF enjoy the property of logical separability. To demonstrate the usefulness of our approach, we apply SDNF in multi-agent epistemic planning. Finally, we extend these results to three more complex multi-agent logics Dn, K45n and KD45n.
AAAI Conference 2019 Conference Paper
The transformed domain fearures of images show effectiveness in distinguishing salient and non-salient regions. In this paper, we propose a novel deep complex neural network, named Sal- DCNN, to predict image saliency by learning features in both pixel and transformed domains. Before proposing Sal-DCNN, we analyze the saliency cues encoded in discrete Fourier transform (DFT) domain. Consequently, we have the following findings: 1) the phase spectrum encodes most saliency cues; 2) a certain pattern of the amplitude spectrum is important for saliency prediction; 3) the transformed domain spectrum is robust to noise and down-sampling for saliency prediction. According to these findings, we develop the structure of Sal- DCNN, including two main stages: the complex dense encoder and three-stream multi-domain decoder. Given the new Sal- DCNN structure, the saliency maps can be predicted under the supervision of ground-truth fixation maps in both pixel and transformed domains. Finally, the experimental results show that our Sal-DCNN method outperforms other 8 state-of-theart methods for image saliency prediction on 3 databases.
NeurIPS Conference 2019 Conference Paper
SARAH and SPIDER are two recently developed stochastic variance-reduced algorithms, and SPIDER has been shown to achieve a near-optimal first-order oracle complexity in smooth nonconvex optimization. However, SPIDER uses an accuracy-dependent stepsize that slows down the convergence in practice, and cannot handle objective functions that involve nonsmooth regularizers. In this paper, we propose SpiderBoost as an improved scheme, which allows to use a much larger constant-level stepsize while maintaining the same near-optimal oracle complexity, and can be extended with proximal mapping to handle composite optimization (which is nonsmooth and nonconvex) with provable convergence guarantee. In particular, we show that proximal SpiderBoost achieves an oracle complexity of O(min{n^{1/2}\epsilon^{-2}, \epsilon^{-3}}) in composite nonconvex optimization, improving the state-of-the-art result by a factor of O(min{n^{1/6}, \epsilon^{-1/3}}). We further develop a novel momentum scheme to accelerate SpiderBoost for composite optimization, which achieves the near-optimal oracle complexity in theory and substantial improvement in experiments.
EAAI Journal 2019 Journal Article
NeurIPS Conference 2018 Conference Paper
Cubic-regularized Newton's method (CR) is a popular algorithm that guarantees to produce a second-order stationary solution for solving nonconvex optimization problems. However, existing understandings of convergence rate of CR are conditioned on special types of geometrical properties of the objective function. In this paper, we explore the asymptotic convergence rate of CR by exploiting the ubiquitous Kurdyka-Lojasiewicz (KL) property of the nonconvex objective functions. In specific, we characterize the asymptotic convergence rate of various types of optimality measures for CR including function value gap, variable distance gap, gradient norm and least eigenvalue of the Hessian matrix. Our results fully characterize the diverse convergence behaviors of these optimality measures in the full parameter regime of the KL property. Moreover, we show that the obtained asymptotic convergence rates of CR are order-wise faster than those of first-order gradient descent algorithms under the KL property.
AAAI Conference 2018 Conference Paper
Existential rules, a family of expressive ontology languages, inherit desired expressive and reasoning properties from both description logics and logic programming. On the other hand, forgetting is a well studied operation for ontology reuse, obfuscation and analysis. Yet it is challenging to establish a theory of forgetting for existential rules. In this paper, we lay the foundation for a theory of forgetting for existential rules by developing a novel notion of unfolding. In particular, we introduce a definition of forgetting for existential rules in terms of query answering and provide a characterisation of forgetting by the unfolding. A result of forgetting may not be expressible in existential rules, and we then capture the expressibility of forgetting by a variant of boundedness. While the expressibility is undecidable in general, we identify a decidable fragment. Finally, we provide an algorithm for forgetting in this fragment.
AAAI Conference 2018 Conference Paper
With the recent advances of neural models and natural language processing, automatic generation of classical Chinese poetry has drawn significant attention due to its artistic and cultural value. Previous works mainly focus on generating poetry given keywords or other text information, while visual inspirations for poetry have been rarely explored. Generating poetry from images is much more challenging than generating poetry from text, since images contain very rich visual information which cannot be described completely using several keywords, and a good poem should convey the image accurately. In this paper, we propose a memory based neural model which exploits images to generate poems. Specifically, an Encoder-Decoder model with a topic memory network is proposed to generate classical Chinese poetry from images. To the best of our knowledge, this is the first work attempting to generate classical Chinese poetry from images with neural networks. A comprehensive experimental investigation with both human evaluation and quantitative analysis demonstrates that the proposed model can generate poems which convey images accurately.
KR Conference 2018 Short Paper
AAAI Conference 2018 Conference Paper
The pattern satisfiability is a fundamental problem for SPARQL. This paper provides a complete analysis of decidability/undecidability of satisfiability problems for SPARQL 1. 1 patterns. A surprising result is the undecidability of satis- fiability for SPARQL 1. 1 patterns when only AND and MI- NUS are expressible. Also, it is shown that any fragment of SPARQL 1. 1 without expressing both AND and MINUS is decidable. These results provide a guideline for future SPARQL query language design and implementation.
IJCAI Conference 2018 Conference Paper
We study the problem of learning first-order rules from large Knowledge Graphs (KGs). With recent advancement in information extraction, vast data repositories in the KG format have been obtained such as Freebase and YAGO. However, traditional techniques for rule learning are not scalable for KGs. This paper presents a new approach RLvLR to learning rules from KGs by using the technique of embedding in representation learning together with a new sampling method. Experimental results show that our system outperforms some state-of-the-art systems. Specifically, for massive KGs with hundreds of predicates and over 10M facts, RLvLR is much faster and can learn much more quality rules than major systems for rule learning in KGs such as AMIE+. We also used the RLvLR-mined rules in an inference module to carry out the link prediction task. In this task, RLvLR outperformed Neural LP, a state-of-the-art link prediction system, in both runtime and accuracy.
EAAI Journal 2017 Journal Article
AAAI Conference 2016 Conference Paper
Hashing techniques are powerful for approximate nearest neighbour (ANN) search. Existing quantization methods in hashing are all focused on scalar quantization (SQ) which is inferior in utilizing the inherent data distribution. In this paper, we propose a novel vector quantization (VQ) method named affinity preserving quantization (APQ) to improve the quantization quality of projection values, which has significantly boosted the performance of state-of-the-art hashing techniques. In particular, our method incorporates the neighbourhood structure in the pre- and post-projection data space into vector quantization. APQ minimizes the quantization errors of projection values as well as the loss of affinity property of original space. An effective algorithm has been proposed to solve the joint optimization problem in APQ, and the extension to larger binary codes has been resolved by applying product quantization to APQ. Extensive experiments have shown that APQ consistently outperforms the state-ofthe-art quantization methods, and has significantly improved the performance of various hashing techniques.
JAIR Journal 2016 Journal Article
Two essential tasks in managing description logic knowledge bases are eliminating problematic axioms and incorporating newly formed ones. Such elimination and incorporation are formalised as the operations of contraction and revision in belief change. In this paper, we deal with contraction and revision for the DL-Lite family through a model-theoretic approach. Standard description logic semantics yields an infinite number of models for DL-Lite knowledge bases, thus it is difficult to develop algorithms for contraction and revision that involve DL models. The key to our approach is the introduction of an alternative semantics called type semantics which can replace the standard semantics in characterising the standard inference tasks of DL-Lite. Type semantics has several advantages over the standard one. It is more succinct and importantly, with a finite signature, the semantics always yields a finite number of models. We then define model-based contraction and revision functions for DL-Lite knowledge bases under type semantics and provide representation theorems for them. Finally, the finiteness and succinctness of type semantics allow us to develop tractable algorithms for instantiating the functions.
IJCAI Conference 2016 Conference Paper
A disjunctive logic program under the answer set semantics can be equivalently translated to a normal logic program by the shifting transformation, if the program is head-cycle-free. In this paper, we provide an answer-set-preserving rewriting of a general disjunctive program to a normal program by first applying the unfolding transformation on atoms that prevent the program from being head-cycle-free, then shifting the resulting program. Different from other transformations that eliminate disjunctions in answer set programming, the new rewriting is efficient for "almost" head-cycle-free programs, i. e. , programs that have only a few atoms that prevent them to be head-cycle-free. Based on the new rewriting, we provide an anytime algorithm to compute answer sets of a disjunctive program by calling solvers for normal logic programs. The experiment shows that the algorithm is efficient for some disjunctive programs. We also extend the rewriting to non-ground answer set programs on finite structures.
EAAI Journal 2016 Journal Article
IJCAI Conference 2016 Conference Paper
We present a novel approach called Minimal Reconstruction Bias Hashing (MRH) to learn similarity preserving binary codes that jointly optimize both projection and quantization stages. Our work tackles an important problem of how to elegantly connect optimizing projection with optimizing quantization, and to maximize the complementary effects of two stages. Distinct from previous works, MRH can adaptively adjust the projection dimensionality to balance the information loss between projection and quantization. It is formulated as a problem of minimizing reconstruction bias of compressed signals. Extensive experiment results have shown the proposed MRH significantly outperforms a variety of state-of-the-art methods over several widely used benchmarks.
AAAI Conference 2015 Conference Paper
Model-based approaches provide a semantically well justified way to revise ontologies. However, in general, model-based revision operators are limited due to lack of efficient algorithms and inexpressibility of the revision results. In this paper, we make both theoretical and practical contribution to efficient computation of model-based revisions in DL-Lite. Specifically, we show that maximal approximations of two well-known model-based revisions for DL-LiteR can be computed using a syntactic algorithm. However, such a coincidence of model-based and syntactic approaches does not hold when role functionality axioms are allowed. As a result, we identify conditions that guarantee such a coincidence for DL-LiteFR. Our result shows that both model-based and syntactic revisions can co-exist seamlessly and the advantages of both approaches can be taken in one revision operator. Based on our theoretical results, we develop a graph-based algorithm for the revision operators and thus graph database techniques can be used to compute ontology revisions. Preliminary evaluation results show that the graph-based algorithm can efficiently handle revision of practical ontologies with large data.
IJCAI Conference 2015 Conference Paper
Classic entrenchment-based contraction is not applicable to many useful logics, such as description logics. This is because the semantic construction refers to arbitrary disjunctions of formulas, while many logics do not fully support disjunction. In this paper, we present a new entrenchment-based contraction which does not rely on any logical connectives except conjunction. This contraction is applicable to all fragments of first-order logic that support conjunction. We provide a representation theorem for the contraction which shows that it satisfies all the AGM postulates except for the controversial Recovery Postulate, and is a natural generalisation of entrenchment-based contraction.
IJCAI Conference 2015 Conference Paper
Hashing is one of the effective techniques for fast Approximate Nearest Neighbour (ANN) search. Traditional single-bit quantization (SBQ) in most hashing methods incurs lots of quantization error which seriously degrades the search performance. To address the limitation of SBQ, researchers have proposed promising multi-bit quantization (MBQ) methods to quantize each projection dimension with multiple bits. However, some MBQ methods need to adopt specific distance for binary code matching instead of the original Hamming distance, which would significantly decrease the retrieval speed. Two typical MBQ methods Hierarchical Quantization and Double Bit Quantization retain the Hamming distance, but both of them only consider the projection dimensions during quantization, ignoring the neighborhood structure of raw data inherent in Euclidean space. In this paper, we propose a multi-bit quantization method named Hamming Compatible Quantization (HCQ) to preserve the capability of similarity metric between Euclidean space and Hamming space by utilizing the neighborhood structure of raw data. Extensive experiment results have shown our approach significantly improves the performance of various stateof-the-art hashing methods while maintaining fast retrieval speed.
AAAI Conference 2015 Conference Paper
The development and maintenance of large and complex ontologies are often time-consuming and error-prone. Thus, automated ontology learning and evolution have attracted intensive research interest. In data-centric applications where ontologies are designed from the data or automatically learnt from it, when new data instances are added that contradict the ontology, it is often desirable to incrementally revise the ontology according to the added data. In description logics, this problem can be intuitively formulated as the operation of TBox contraction, i. e. , rational elimination of certain axioms from the logical consequences of a TBox, and it is w. r. t. an ABox. In this paper we introduce a model-theoretic approach to such a contraction problem by using an alternative semantic characterisation of DL-Lite TBoxes. We show that entailment checking (without necessarily first computing the contraction result) is in coNP, which does not shift the corresponding complexity in propositional logic, and the problem is tractable when the size of the new data is bounded.
AAAI Conference 2015 Conference Paper
The theory of (variable) forgetting has received significant attention in nonmonotonic reasoning, especially, in answer set programming. However, the problem of establishing a theory of forgetting for some expressive nonmonotonic logics such as McCarthy’s circumscription is rarely explored. In this paper a theory of forgetting for propositional circumscription is proposed, which is not a straightforward adaption of existing approaches. In particular, some properties that are essential for existing proposals do not hold any longer or have to be reformulated. Several useful properties of the new forgetting are proved, which demonstrate suitability of the forgetting for circumscription. A sound and complete algorithm for the forgetting is developed and an analysis of computational complexity is given.
AAAI Conference 2015 Conference Paper
AAAI Conference 2014 Conference Paper
Two essential tasks in managing Description Logic (DL) ontologies are eliminating problematic axioms and incorporating newly formed axioms. Such elimination and incorporation are formalised as the operations of contraction and revision in belief change. In this paper, we deal with contraction and revision for the DL-Lite family through a model-theoretic approach. Standard DL semantics yields infinite numbers of models for DL-Lite TBoxes, thus it is not practical to develop algorithms for contraction and revision that involve DL models. The key to our approach is the introduction of an alternative semantics called type semantics which is more succinct than DL semantics. More importantly, with a finite signature, type semantics always yields finite humber of models. We then define model-based contraction and revision for DL-Lite TBoxes under type semantics and provide representation theorems for them. Finally, the succinctness of type semantics allows us to develop tractable algorithms for both operations.
KR Conference 2012 Conference Paper
problem in both database and KR settings. This problem is undecidable (Beeri and Vardi 1981) in general, and it can be characterised using chase (Johnson and Klug 1984; Maier, Mendelzon, and Sagiv 1979), a technique closely related to the hypertableau calculus (Motik, Shearer, and Horrocks 2009). The chase extends in a forward-chaining manner the original set of facts by introducing facts implied by the rules. The result of the chase is called the universal model, and an arbitrary conjunctive query can be answered over the original set of facts and the rules by simply evaluating the query in the universal model. Rules with existentially quantified variables in the head— so-called generating rules—require the introduction of fresh individuals, and cyclic applications of generating rules may lead to non-termination; moreover, determining whether chase terminates on a set of rules and facts is undecidable. However, several decidable classes of existential rules have been identified, and the existing proposals can be classified into two main groups. In the first group, rules are restricted such that their (possibly infinite) universal models can be represented using finitary means. This group includes rules with universal models of bounded treewidth (Baget et al. 2011), guarded rules (Calı̀ et al. 2010), and ‘sticky’ rules (Calı̀, Gottlob, and Pieris 2011). In the second group, one uses a sufficient (but not necessary) acyclicity condition that ensures chase termination. Roughly speaking, acyclicity conditions analyse information flow between the rules to ensure that no cyclic applications of generating rules are possible. Weak acyclicity (WA) (Fagin et al. 2005) was one of the first such notions, and it was extended to safety (SF) (Meier, Schmidt, and Lausen 2009), stratification (ST) (Deutsch, Nash, and Remmel 2008), acyclicity of a graph of rule dependencies (aGRD) (Baget, Mugnier, and Thomazo 2011), joint acyclicity (JA) (Krötzsch and Rudolph 2011), and super-weak acyclicity (SWA) (Marnette 2009). Acyclicity conditions are relevant for at least two reasons. First, unlike guarded rules, acyclic rules can axiomatise structures of arbitrary shapes, as long as these structures are bounded in size. Second, the chase result for acyclic rules can be stored and manipulated as if it were a database. This is important in data exchange, where the goal is to materialise the transformed database. In this paper, we argue that acyclicity is also relevant for description logics (DLs), the KR formalisms underpin- Answering conjunctive queries (CQs) over a set of facts extended with existential rules is a key problem in knowledge representation and databases. This problem can be solved using the chase (aka materialisation) algorithm; however, CQ answering is undecidable for general existential rules, so the chase is not guaranteed to terminate. Several acyclicity conditions provide sufficient conditions for chase termination. In this paper, we present two novel such conditions—modelfaithful acyclicity (MFA) and model-summarising acyclicity (MSA)—that generalise many of the acyclicity conditions known so far in the literature. Materialisation provides the basis for several widely-used OWL 2 DL reasoners. In order to avoid termination problems, many of these systems handle only the OWL 2 RL profile of OWL 2 DL; furthermore, some systems go beyond OWL 2 RL, but they provide no termination guarantees. In this paper we investigate whether various acyclicity conditions can provide a principled and practical solution to these problems. On the theoretical side, we show that query answering for acyclic ontologies is of lower complexity than for general ontologies. On the practical side, we show that many of the commonly used OWL 2 DL ontologies are MSA, and that the facts obtained via materialisation are not too large. Thus, our results suggest that principled extensions to materialisationbased OWL 2 DL reasoners may be practically feasible.
AAAI Conference 2010 Conference Paper
Revising knowledge bases (KBs) in description logics (DLs) in a syntax-independent manner is an important, nontrivial problem for the ontology management and DL communities. Several attempts have been made to adapt classical modelbased belief revision and update techniques to DLs, but they are restricted in several ways. In particular, they do not provide operators or algorithms for general DL KB revision. The key difficulty is that, unlike propositional logic, a DL KB may have infinitely many models with complex (and possibly infinite) structures, making it difficult to define and compute revisions in terms of models. In this paper, we study general KBs in a specific DL in the DL-Lite family. We introduce the concept of features for such KBs, develop an alternative semantic characterization of KBs using features (instead of models), define two specific revision operators for KBs, and present the first algorithm for computing best approximations for syntax-independent revisions of KBs.
KR Conference 2010 Conference Paper
This paper introduces a new methodology of revising general KBs in DL-Lite. Two specific revision operators are defined, their properties are investigated and algorithms for computing revisions are developed.
IROS Conference 2006 Conference Paper
A new server-decentralized Internet architecture based on Jabber for robot remote control is proposed in order to have high availability, performance, scalability, fault tolerance and security. Four components of the architecture: operators, robots, transfer servers and data-keeper are defined and functioned. The robot-controlling data/robot state data are packed with XML stanzas and delivered to the addressed robot/operator through XML streams. In order to test its availability, the architecture is implemented and instanced as a remote control simulation system of Puma560 robot. The system is experimented and a network test is carried out to evaluate the structure; the results show that the architecture is suitable for many kinds of robot remote control scenarios despite the tough network environment