IJCAI 2024 Papers

In this paper, we present the Sub-Adjacent Transformer with a novel attention mechanism for unsupervised time series anomaly detection. Unlike previous approaches that rely on all the points within some neighborhood for time point reconstruction, our method restricts the attention to regions not immediately adjacent to the target points, termed sub-adjacent neighborhoods. Our key observation is that owing to the rarity of anomalies, they typically exhibit more pronounced differences from their sub-adjacent neighborhoods than from their immediate vicinities. By focusing the attention on the sub-adjacent areas, we make the reconstruction of anomalies more challenging, thereby enhancing their detectability. Technically, our approach concentrates attention on the non-diagonal areas of the attention matrix by enlarging the corresponding elements in the training stage. To facilitate the implementation of the desired attention matrix pattern, we adopt linear attention because of its flexibility and adaptability. Moreover, a learnable mapping function is proposed to improve the performance of linear attention. Empirically, the Sub-Adjacent Transformer achieves state-of-the-art performance across six real-world anomaly detection benchmarks, covering diverse fields such as server monitoring, space exploration, and water treatment.

Both structured d-DNNF and SDD can be exponentially more succinct than OBDD. Moreover, SDD is essentially as tractable as OBDD. But this leaves left two important open questions. Firstly, does OBDD support more tractable transformations than structured d-DNNF? And secondly, is structured d-DNNF more succinct than SDD? In this paper, we answer both questions in the affirmative. For the first question we show that, unlike OBDD, structured d-DNNF does not support polytime negation, disjunction, or existential quantification operations. As a corollary, we deduce that there are functions with an equivalent polynomial-sized structured d-DNNF but with no such representation as an SDD, thus answering the second question. We also lift this second result to arithmetic circuits (AC) to show a succinctness gap between PSDD and the positive AC analogue to structured d-DNNF.

Recently, there has been growing interest in using physics-informed neural networks (PINNs) to solve differential equations. However, the preservation of structure, such as energy and stability, in a suitable manner has yet to be established. This limitation could be a potential reason why the learning process for PINNs is not always efficient and the numerical results may suggest nonphysical behavior. Besides, there is little research on their applications on downstream tasks. To address these issues, we propose structure-preserving PINNs to improve their performance and broaden their applications for downstream tasks. Firstly, by leveraging prior knowledge about the physical system, a structure‐preserving loss function is designed to assist the PINN in learning the underlying structure. Secondly, a framework that utilizes structure-preserving PINN for robust image recognition is proposed. Here, preserving the Lyapunov structure of the underlying system ensures the stability of the system. Experimental results demonstrate that the proposed method improves the numerical accuracy of PINNs for partial differential equations (PDEs). Furthermore, the robustness of the model against adversarial perturbations in image data is enhanced.

Video shadow detection faces significant challenges due to ambiguous semantics and variable shapes. Existing video shadow detection algorithms typically overlook the fine shadow details, resulting in inconsistent detection between consecutive frames in complex real-world video scenarios. To address this issue, we propose a spatial-temporal feature interaction strategy, which refines and enhances global shadow semantics with local prior features in the modeling of shadow relations between frames. Moreover, a structure-aware shadow prediction module is proposed, which focuses on modeling the distance relation between local shadow edges and regions. Quantitative experimental results demonstrate that our approach significantly outperforms the state-of-the-art methods, providing stable and consistent shadow detection results in complex video shadow scenarios.

In recent years, employing layer attention to enhance interaction among hierarchical layers has proven to be a significant advancement in building network structures. In this paper, we delve into the distinction between layer attention and the general attention mechanism, noting that existing layer attention methods achieve layer interaction on fixed feature maps in a static manner. These static layer attention methods limit the ability for context feature extraction among layers. To restore the dynamic context representation capability of the attention mechanism, we propose a Dynamic Layer Attention (DLA) architecture. The DLA comprises dual paths, where the forward path utilizes an improved recurrent neural network block, named Dynamic Sharing Unit (DSU), for context feature extraction. The backward path updates features using these shared context representations. Finally, the attention mechanism is applied to these dynamically refreshed feature maps among layers. Experimental results demonstrate the effectiveness of the proposed DLA architecture, outperforming other state-of-the-art methods in image recognition and object detection tasks. Additionally, the DSU block has been evaluated as an efficient plugin in the proposed DLA architecture. The code is available at https://github.com/tunantu/Dynamic-Layer-attention.

Input/Output (I/O) logic is a general framework for reasoning about conditional norms and/or causal relations. We streamline Bochman’s causal I/O logics and their original version via proof-search-oriented sequent calculi. As a byproduct, we obtain new, simple semantics for all these logics, complexity bounds, embeddings into normal modal logics, and efficient deduction methods. Our work encompasses many scattered results and provides uniform solutions to various unresolved problems.

Predicting the spatiotemporal variation in streamflow along with uncertainty quantification enables decision-making for sustainable management of scarce water resources. Process-based hydrological models (aka physics-based models) are based on physical laws, but use simplifying assumptions which can lead to poor accuracy. Data-driven approaches offer a powerful alternative, but they require large amount of training data and tend to produce predictions that are inconsistent with physical laws. This paper studies a constrained reasoning and learning (CRL) approach where physical laws represented as logical constraints are integrated as a layer in the deep neural network. To address small data setting, we develop a theoretically-grounded training approach to improve the generalization accuracy of deep models. For uncertainty quantification, we combine the synergistic strengths of Gaussian processes (GPs) and deep temporal models by passing the learned latent representation as input to a standard distance-based kernel. Experiments on multiple real-world datasets demonstrate the effectiveness of both CRL and GP with deep kernel approaches over strong baseline methods.

Matching markets consist of two disjoint sets of agents, where each agent has a preference list over agents on the other side. The primary objective is to find a stable matching between the agents such that no unmatched pair of agents prefer each other to their matched partners. The incompatibility between stability and strategy-proofness in this domain gives rise to a variety of strategic behavior of agents, which in turn may influence the resulting matching. In this paper, we discuss fundamental properties of stable matchings, review essential structural observations, survey key results in manipulation algorithms and their game-theoretical aspects, and more importantly, highlight a series of open research questions.

Neural simulators for modeling complex dynamical systems have been extensively studied for various real-world applications, such as weather forecasting, ocean current prediction, and computational fluid dynamics simulation. Although they have demonstrated powerful fitting and predicting, most existing models are only built to learn single-system dynamics. Several advanced researches have considered learning dynamics across environments, which can exploit the potential commonalities among the dynamics across environments and adapt to new environments. However, these methods still are prone to scarcity problems where per-environment data is sparse or limited. Therefore, we propose a novel CoNDP (Context-Informed Neural ODE Processes) to achieve learning system dynamics from sparse observations across environments. It can fully use contextual information of each environment to better capture the intrinsic commonalities across environments and distinguishable differences among environments while modeling uncertainty of system evolution, producing more accurate predictions. Intensive experiments are conducted on five complex dynamical systems in various fields. Results show that the proposed CoNDP can achieve optimal results compared with common neural simulators and state-of-the-art cross-environmental models.

Auction-based federated learning (AFL) is an important area of FL incentive mechanism design. It effectively incentivizes high-quality data owners (DOs) to participate in data consumers' (DCs, i.e., servers') FL training tasks. However, AFL is still evolving, with existing methods primarily addressing optimal DC-DO matching or DC selection problems in monopoly markets. To enhance the practicality of AFL, we introduce stakeholder-oriented decision support in AFL. This facilitates optimal and strategic decision-making for all stakeholders, improving the efficiency and sustainability of the AFL ecosystem.

Detecting synthetic from real speech is increasingly crucial due to the risks of misinformation and identity impersonation. While various datasets for synthetic speech analysis have been developed, they often focus on specific areas, limiting their utility for comprehensive research. To fill this gap, we propose the Speech-Forensics dataset by extensively covering authentic, synthetic, and partially forged speech samples that include multiple segments synthesized by different high-quality algorithms. Moreover, we propose a TEmporal Speech LocalizaTion network, called TEST, aiming at simultaneously performing authenticity detection, multiple fake segments localization, and synthesis algorithms recognition, without any complex post-processing. TEST effectively integrates LSTM and Transformer to extract more powerful temporal speech representations and utilizes dense prediction on multi-scale pyramid features to estimate the synthetic spans. Our model achieves an average mAP of 83.55% and an EER of 5.25% at the utterance level. At the segment level, it attains an EER of 1.07% and a 92.19% F1 score. These results highlight the model's robust capability for a comprehensive analysis of synthetic speech, offering a promising avenue for future research and practical applications in this field.

Representing temporal-structured samples is essential for effective time series analysis tasks. So far, recurrent networks, convolution networks and transformer-style models have been successively applied in temporal data representation, yielding notable results. However, most existing methods primarily focus on modeling and representing the variation patterns within time series in the time domain. As a highly abstracted information entity, 1D time series couples various patterns such as trends, seasonality, and dramatic changes (instantaneous high dynamic), it is difficult to exploit these highly coupled properties merely by analysis tools on purely time domain. To this end, we present Spectrogram Analysis and Representation Network (SpecAR-Net). SpecAR-Net aims at learning more comprehensive representations by modeling raw time series in both time and frequency domain, where an efficient joint extraction of time-frequency features is achieved through a group of learnable 2D multi-scale parallel complex convolution blocks. Experimental results show that the SpecAR-Net achieves excellent performance on 5 major downstream tasks i.e., classification, anomaly detection, imputation, long- and short-term forecasting. Code and appendix are available at https://github.com/Dongyi2go/SpecAR_Net.

As Artificial Intelligence evolves, the neural models vulnerable to adversarial attacks may produce fatal results in critical applications. This paper mainly discusses the robustness of the compressed neural models facing adversarial attacks. A few studies discuss the interaction between model compression and adversarial attack. However, they focus on the robustness against the traditional attacks designed for the dense models, not the attacks intended explicitly for the compressed models, using sparsity and quantization techniques. Compressed models often have fewer parameters and smaller sizes that are more friendly to resource-limited devices than dense models, so they are widely deployed in various edge and mobile devices. However, introducing the sparsity and quantization into neural models further imposes higher attack risks. A specific adversarial attack method (Spear) is proposed to generate the particular adversarial attack samples for evaluating the robustness of the compressed models. The Spear attack finds minimal perturbations to create the attack samples to maximize the different behaviors between the compressed and dense reference models. We demonstrate the proposed Spear attack technique can generally be applied to various networks and tasks through quantitative and ablation experiments.

The air quality inference problem aims to utilize historical data from a limited number of observation sites to infer the air quality index at an unknown location. Considering the sparsity of data due to the high maintenance cost of the stations, good inference algorithms can effectively save the cost and refine the data granularity. While spatio-temporal graph neural networks have made excellent progress on this problem, their non-Euclidean and discrete data structure modeling of reality limits its potential. In this work, we make the first attempt to combine two different spatio-temporal perspectives, fields and graphs, by proposing a new model, Spatio-Temporal Field Neural Network, and its corresponding new framework, Pyramidal Inference. Extensive experiments validate that our model achieves state-of-the-art performance in nationwide air quality inference in the Chinese Mainland, demonstrating the superiority of our proposed model and framework.

Spatiotemporal forecasting techniques are significant for various domains such as transportation, energy, and weather. Accurate prediction of spatiotemporal series remains challenging due to the complex spatiotemporal heterogeneity. In particular, current end-to-end models are limited by input length and thus often fall into spatiotemporal mirage, i.e., similar input time series followed by dissimilar future values and vice versa. To address these problems, we propose a novel self-supervised pre-training framework Spatial-Temporal-Decoupled Masked Pre-training (STD-MAE) that employs two decoupled masked autoencoders to reconstruct spatiotemporal series along the spatial and temporal dimensions. Rich-context representations learned through such reconstruction could be seamlessly integrated by downstream predictors with arbitrary architectures to augment their performances. A series of quantitative and qualitative evaluations on four widely used benchmarks (PEMS03, PEMS04, PEMS07, and PEMS08) are conducted to validate the state-of-the-art performance of STD-MAE. Codes are available at https://github.com/Jimmy-7664/STD-MAE.

Session-based recommendation (SBR) aims to predict the next-interacted item based on anonymous users' behavior sequences. The main challenge is how to recognize the user intent with limited interactions to achieve a more accurate inference of user behavior. Existing works usually regard several consecutive items in the current session as intent. However, we argue such intent generation based on temporal transition ignores the fact that each item also has its semantically connected items in the feature space, which can be regarded as spatial intent. The limited consideration of intent fails to capture complex behavioral patterns in real-world scenarios, leading to sub-optimal solutions. To address this issue, we propose the Hierarchical Intent Perceiving Contrastive Learning Framework (HearInt) for SBR, which proposes a hierarchical consideration of intents from both temporal and spatial perspective. Specifically, we first propose that the user's temporal intents are mutually exclusive while the spatial intents are mutually compatible. Following these analyses, we design a Temporal Intent Decoupling module to mitigate the mutual influence of long-term and short-term intents, and a Cross-scale Contrastive Learning task to enhance the consistency of intents across different spatial scales. Experimental results on three real-world datasets exhibit that HearInt achieves state-of-the-art performance.

Object trajectory prediction is a hot research issue with wide applications in video surveillance and autonomous driving. The previous studies consider the interaction sparsity mainly among the pedestrians instead of multi-type of objects, which brings new types of interactions and consequently superfluous ones. This paper proposes a Multi-type Object Trajectory Prediction (MOTP) method with a Sparse Multi-relational Graph Convolutional Network (SMGCN) and a novel multi-round Global Temporal Aggregation (GTA). MOTP introduces a novel adaptive sparsification and multi-scale division method to model interactions among multitype of objects. It further incorporates a Sparse Multi-relational Temporal Graph to capture the temporal division of multi-type trajectories, along with a multi-round Global Temporal Aggregation (GTA) mechanism to mitigate error accumulation, and enhances the trajectory prediction accuracy. The extensive evaluation on the ETH, UCY and SDD datasets shows that our method outperforms the typical state-of-the-art works by significant margins. Codes will be available in https://github.com/ sounio/SMGCN.

Traditional Named Entity Recognition (NER) models are typically designed for domain-specific datasets and limited to fixed predefined types, resulting in difficulty generalizing to new domains. Recently, prompt-based generative methods attempt to mitigate this constraint by training models jointly on diverse datasets and extract specified entities via prompt instructions. However, due to autoregressive structure, these methods cannot directly model entity span and suffer from slow sequential decoding. To address these issues, we propose a novel Span-based Unified NER framework via contrastive learning (SUNER), which aligns text span and entity type representations in a shared semantic space to extract entities in parallel. Specifically, we first extract mention spans without considering entity types to better generalize across datasets. Then, by leveraging the power of contrastive learning and well-designed entity marker structure, we map candidate spans and their textual type descriptions into the same vector representation space to differentiate entities across domains. Extensive experiments on both supervised and zero/few-shot settings demonstrate that proposed SUNER model achieves better performance and higher efficiency than previous state-of-the-art unified NER models.

The quantified Boolean formula (QBF) problem is an important decision problem generally viewed as the archetype for PSPACE-completeness. Many problems of central interest in AI are in general not included in NP, e.g., planning, model checking, and non-monotonic reasoning, and for such problems QBF has successfully been used as a modelling tool. However, solvers for QBF are not as advanced as state of the art SAT solvers, which has prevented QBF from becoming a universal modelling language for PSPACE-complete problems. A theoretical explanation is that QBF (as well as many other PSPACE-complete problems) lacks natural parameters guaranteeing fixed-parameter tractability (FPT). In this paper we tackle this problem and consider a simple but overlooked parameter: the number of existentially quantified variables. This natural parameter is virtually unexplored in the literature which one might find surprising given the general scarcity of FPT algorithms for QBF. Via this parameterization we then develop a novel FPT algorithm applicable to QBF instances in conjunctive normal form (CNF) of bounded clause length. We complement this by a W[1]-hardness result for QBF in CNF of unbounded clause length as well as sharper lower bounds for the bounded arity case under the (strong) exponential-time hypothesis.