Temporal knowledge graph (TKG) reasoning aims to predict the future missing facts based on historical information and has gained increasing research interest recently. Lots of works have been made to model the historical structural and temporal characteristics for the reasoning task. Most existing works model the graph structure mainly depending on entity representation. However, the magnitude of TKG entities in real-world scenarios is considerable, and an increasing number of new entities will arise as time goes on. Therefore, we propose a novel architecture modeling with relation feature of TKG, namely aDAptivE path-MemOry Network (DaeMon), which adaptively models the temporal path information between query subject and each object candidate across history time. It models the historical information without depending on entity representation. Specifically, DaeMon uses path memory to record the temporal path information derived from path aggregation unit across timeline considering the memory passing strategy between adjacent timestamps. Extensive experiments conducted on four real-world TKG datasets demonstrate that our proposed model obtains substantial performance improvement and outperforms the state-of-the-art up to 4.8% absolute in MRR.
Recently, Visual Information Extraction (VIE) has been becoming increasingly important in both the academia and industry, due to the wide range of real-world applications. Previously, numerous works have been proposed to tackle this problem. However, the benchmarks used to assess these methods are relatively plain, i.e., scenarios with real-world complexity are not fully represented in these benchmarks. As the first contribution of this work, we curate and release a new dataset for VIE, in which the document images are much more challenging in that they are taken from real applications, and difficulties such as blur, partial occlusion, and printing shift are quite common. All these factors may lead to failures in information extraction. Therefore, as the second contribution, we explore an alternative approach to precisely and robustly extract key information from document images under such tough conditions. Specifically, in contrast to previous methods, which usually either incorporate visual information into a multi-modal architecture or train text spotting and information extraction in an end-to-end fashion, we explicitly model entities as semantic points, i.e., center points of entities are enriched with semantic information describing the attributes and relationships of different entities, which could largely benefit entity labeling and linking. Extensive experiments on standard benchmarks in this field as well as the proposed dataset demonstrate that the proposed method can achieve significantly enhanced performance on entity labeling and linking, compared with previous state-of-the-art models. Dataset is available at https://www.modelscope.cn/datasets/damo/SIBR/summary.
The goal of domain generalization (DG) is to enhance the generalization capability of the model learned from a source domain to other unseen domains. The recently developed Sharpness-Aware Minimization (SAM) method aims to achieve this goal by minimizing the sharpness measure of the loss landscape. Though SAM and its variants have demonstrated impressive DG performance, they may not always converge to the desired flat region with a small loss value. In this paper, we present two conditions to ensure that the model could converge to a flat minimum with a small loss, and present an algorithm, named Sharpness-Aware Gradient Matching (SAGM), to meet the two conditions for improving model generalization capability. Specifically, the optimization objective of SAGM will simultaneously minimize the empirical risk, the perturbed loss (i.e., the maximum loss within a neighborhood in the parameter space), and the gap between them. By implicitly aligning the gradient directions between the empirical risk and the perturbed loss, SAGM improves the generalization capability over SAM and its variants without increasing the computational cost. Extensive experimental results show that our proposed SAGM method consistently outperforms the state-of-the-art methods on five DG benchmarks, including PACS, VLCS, OfficeHome, TerraIncognita, and DomainNet. Codes are available at https://github.com/Wang-pengfei/SAGM.
Feature Selection (FS), such as filter, wrapper, and embedded methods, aims to find the optimal feature subset for a given downstream task. However, in many real-world practices, 1) the criteria of FS vary across domains; 2) FS is brittle when data is a high-dimensional and small sample size. Can selected feature subsets be more generalized, accurate, and input dimensionality agnostic? We generalize this problem into a deep differentiable feature selection task and propose a new perspective: discrete feature subsetting as continuous embedding space optimization. We develop a generic and principled framework including a deep feature subset encoder, accuracy evaluator, decoder, and gradient ascent optimizer. This framework implements four steps: 1) features-accuracy training data preparation; 2) deep feature subset embedding; 3) gradient-optimized search; 4) feature subset reconstruction. We develop new technical insights: reinforcement as a training data generator, ensembles of diverse peer and exploratory feature selector knowledge for generalization, an effective embedding from feature subsets to continuous space along with joint optimizing reconstruction and accuracy losses to select accurate features. Experimental results demonstrate the effectiveness of the proposed method.
Non-autoregressive translation (NAT) model achieves a much faster inference speed than the autoregressive translation (AT) model because it can simultaneously predict all tokens during inference. However, its translation quality suffers from degradation compared to AT. And existing NAT methods only focus on improving the NAT model's performance but do not fully utilize it. In this paper, we propose a simple but effective method called "Candidate Soups," which can obtain high-quality translations while maintaining the inference speed of NAT models. Unlike previous approaches that pick the individual result and discard the remainders, Candidate Soups (CDS) can fully use the valuable information in the different candidate translations through model uncertainty. Extensive experiments on two benchmarks (WMT'14 EN-DE and WMT'16 EN-RO) demonstrate the effectiveness and generality of our proposed method, which can significantly improve the translation quality of various base models. More notably, our best variant outperforms the AT model on three translation tasks with 7.6 times speedup.
Feature transformation for AI is an essential task to boost the effectiveness and interpretability of machine learning (ML). Feature transformation aims to transform original data to identify an optimal feature space that enhances the performances of a downstream ML model. Existing studies either combines preprocessing, feature selection, and generation skills to empirically transform data, or automate feature transformation by machine intelligence, such as reinforcement learning. However, existing studies suffer from: 1) high-dimensional non-discriminative feature space; 2) inability to represent complex situational states; 3) inefficiency in integrating local and global feature information. To fill the research gap, we formulate the feature transformation task as an iterative, nested process of feature generation and selection, where feature generation is to generate and add new features based on original features, and feature selection is to remove redundant features to control the size of feature space. Finally, we present extensive experiments and case studies to illustrate 24.7\% improvements in F1 scores compared with SOTAs and robustness in high-dimensional data.
Urban traffic speed prediction aims to estimate the future traffic speed for improving the urban transportation services. Enormous efforts have been made on exploiting spatial correlations and temporal dependencies of traffic speed evolving patterns by leveraging explicit spatial relations (geographical proximity) through pre-defined geographical structures ({\it e.g.}, region grids or road networks). While achieving promising results, current traffic speed prediction methods still suffer from ignoring implicit spatial correlations (interactions), which cannot be captured by grid/graph convolutions. To tackle the challenge, we propose a generic model for enabling the current traffic speed prediction methods to preserve implicit spatial correlations. Specifically, we first develop a Dual-Transformer architecture, including a Spatial Transformer and a Temporal Transformer. The Spatial Transformer automatically learns the implicit spatial correlations across the road segments beyond the boundary of geographical structures, while the Temporal Transformer aims to capture the dynamic changing patterns of the implicit spatial correlations. Then, to further integrate both explicit and implicit spatial correlations, we propose a distillation-style learning framework, in which the existing traffic speed prediction methods are considered as the teacher model, and the proposed Dual-Transformer architectures are considered as the student model. The extensive experiments over three real-world datasets indicate significant improvements of our proposed framework over the existing methods.
Graph contrastive learning (GCL) has been an emerging solution for graph self-supervised learning. The core principle of GCL is to reduce the distance between samples in the positive view, but increase the distance between samples in the negative view. While achieving promising performances, current GCL methods still suffer from two limitations: (1) uncontrollable validity of augmentation, that graph perturbation may produce invalid views against semantics and feature-topology correspondence of graph data; and (2) unreliable binary contrastive justification, that the positiveness and negativeness of the constructed views are difficult to be determined for non-euclidean graph data. To tackle the above limitations, we propose a new contrastive learning paradigm for graphs, namely Graph Soft-Contrastive Learning (GSCL), that conducts contrastive learning in a finer-granularity via ranking neighborhoods without any augmentations and binary contrastive justification. GSCL is built upon the fundamental assumption of graph proximity that connected neighbors are more similar than far-distant nodes. Specifically, we develop pair-wise and list-wise Gated Ranking infoNCE Loss functions to preserve the relative ranking relationship in the neighborhood. Moreover, as the neighborhood size exponentially expands with more hops considered, we propose neighborhood sampling strategies to improve learning efficiency. The extensive experimental results show that our proposed GSCL can consistently achieve state-of-the-art performances on various public datasets with comparable practical complexity to GCL.
Domain generalization (DG) for person re-identification (ReID) is a challenging problem, as there is no access to target domain data permitted during the training process. Most existing DG ReID methods employ the same features for the updating of the feature extractor and classifier parameters. This common practice causes the model to overfit to existing feature styles in the source domain, resulting in sub-optimal generalization ability on target domains even if meta-learning is used. To solve this problem, we propose a novel style interleaved learning framework. Unlike conventional learning strategies, interleaved learning incorporates two forward propagations and one backward propagation for each iteration. We employ the features of interleaved styles to update the feature extractor and classifiers using different forward propagations, which helps the model avoid overfitting to certain domain styles. In order to fully explore the advantages of style interleaved learning, we further propose a novel feature stylization approach to diversify feature styles. This approach not only mixes the feature styles of multiple training samples, but also samples new and meaningful feature styles from batch-level style distribution. Extensive experimental results show that our model consistently outperforms state-of-the-art methods on large-scale benchmarks for DG ReID, yielding clear advantages in computational efficiency. Code is available at https://github.com/WentaoTan/Interleaved-Learning.