Traffic flow forecasting plays a vital role in improving transportation efficiency, safety, and sustainability in cities. Traditional methods focus on capturing the intricate spatiotemporal dynamics within traffic flows. With the development of web and sensing technology, urban knowledge graphs (KG) emerge as an auxiliary knowledge that represents dependencies among traffic entities. Prior works have highlighted the importance of incorporating knowledge from urban KGs to enhance the accuracy of traffic flow forecasting. Unfortunately, these methods inadequately capture the dynamics within the evolving urban KGs, resulting in sub-optimal traffic forecasts.To bridge this gap, we first perform an analytical study to show the impact of short-term and long-term dynamics of urban KGs on the traffic flow forecasting task. Based on our preliminary observation, we propose Dynamic Knowledge Graph-Informed Traffic Flow Forecasting (DKG-TF2), which is a graph neural network-based spatiotemporal forecasting model, enabling to capture of both static and dynamic characteristics from urban KGs, and integrates it with traffic flow forecasting. Evaluated over a real-world dataset, we demonstrate that the DKG-TF2 model significantly outperforms competitive baselines and achieve a root mean squared error of 3.4923. Further ablation studies reveal the complementary roles of static attributes and dynamic attributes in enhancing forecasting accuracy and model robustness.​

Gait recognition is an emerging remote biometric technology that has applications in various fields such as video surveillance. Silhouette-based approach and skeleton-based approach are the two most popular approaches. And researchers are starting to work on the gait recognition methods with the fusion of the two modalities. Therefore, this article focuses on a skeleton-based gait recognition framework (GENGait) with strong generalization capabilities, laying the foundation for future multimodal fusion. Through previous work we have found that previous studies of skeleton-based methods have devised many skeleton preprocessing methods and used many different backbone networks, but their backbones are usually shallow. However, with the publication of the large gait recognition in the wild (GREW) wild dataset, a deep backbone is critical to the generalisation ability of skeleton-based methods. Moreover, by visualizing and analyzing gait sequences, we found that many methods do not pay attention to the fact that the camera drops frames when taking video, which leads to the introduction of noise in the feature extraction. In view of these problems, this article constructs a generalized skeleton-based deep gait learning framework with strong generalization ability from three aspects: the depth of the backbone network, the preprocessing of skeleton sequences, and the processing of time sequences. The network is trained on the GREW dataset and tested on other datasets to evaluate the generalization ability of the network. Our proposed GENGait achieves better results compared to many state-of-the-art skeleton-based methods.

Learning with softmax cross-entropy on one-hot labels often leads to overconfidence on the correct class. While label smoothing regulates this overconfidence by redistributing some confidence from the correct class to other incorrect classes, it compromises the representation in the logits about the similarity between samples of different classes and may hurt calibration if higher confidence is required for high accuracy. To overcome these limitations, we propose a Virtual Smoothing (VS) label that redistributes certain confidence from the correct class to additional VS classes to regularize overconfidence. In VS labels, the VS class nodes act as adversaries to the original class nodes, enforcing regularization by clustering samples across all classes. The zero confidence assigned to each incorrect class also allows the incorrect logits to be different from each other without erasing information about sample similarities. The prediction probability can still approach 1 when applying softmax to the logits of the original real classes, which avoids harming but consistently improves calibration. Experiments show that VS labels consistently improve accuracy and calibration while providing better logits for improved knowledge distillation. Additionally, VS labels exhibit effectiveness in improving adversarial training, robust distillation, and out-of-distribution detection.

The application of machine learning methods to solve combinatorial problems has garnered considerable research interest. In this paper, we propose MAgg (**M**etamorphic **Agg**regation), a method to augment machine learning models for combinatorial problems at inference time using metamorphic relations. MAgg models metamorphic relations using directed graphs, which are then fed to a Graph Neural Network (GNN) model to improve the aggregation of predictions across transformed input instances. By incorporating metamorphic relations, MAgg essentially extends standard Test-Time Augmentation (TTA), eliminating the necessity of label-preserving transformations and expanding its applicability to a broader range of supervised learning tasks for combinatorial problems. We evaluate the proposed MAgg method on three mainstream machine learning tasks for combinatorial problems, namely Boolean Satisfiability Prediction (SAT), Decision Traveling Salesman Problem Satisfiability Prediction (Decision TSP), and Graph Edit Distance Estimation (GED). The evaluation result shows significant improvements over base models in all three tasks, corroborating the effectiveness and versatility of the proposed method.

Abstract A new trend in long-range biometrics, gait recognition, is finding application in a number of different fields including video surveillance. Recently, with the increase in robustness of the pose estimator and the presence of various unpredictable factors in realistic gait recognition, skeleton-based methods with higher robustness have emerged to better meet the challenging gait recognition needs. However, existing approaches primarily focus on extracting global skeletal features, neglecting the intricate motion information of local body parts and overlooking inter-limb relationships. Our solution to these challenges is the dynamic local fusion network (GaitDLF), a novel gait neural network for complex environments that includes a detail-aware stream in addition to the previous direct extraction of global skeleton features, which provides an enhanced representation of gait features. To extract discriminative local motion information, we introduce predefined body part assignments for each joint in the skeletal structure. By segmenting and mapping the overall skeleton based on these limb site divisions, limb-level motion features can be obtained. In addition, we will dynamically fuse the motion features from different limbs and enhance the motion feature representation of each limb by global context information and local context information of the limb-level motion features. The ability to extract gait features between individuals can be improved by aggregating local motion features from different body parts. Based on experiments on CASIA-B, Gait3D, and GREW, we show that our model extracts more comprehensive gait features than the state-of-the-art skeleton-based method, demonstrating that our method is better suited to detecting gait in complex environments in the wild than the appearance-based method.

Parallelization is a promising way to improve the performance of Python programs. Unfortunately, developers may miss parallelization possibilities, because they usually do not concentrate on parallelization. Many approaches have been proposed to parallelize Python programs automatically, however, they are either domain-specific or require manual annotation. Thus they cannot solve the problem well in general. In this paper, we propose PyPar, an effective tool aiming at discovering parallelization possibilities in real-world Python programs. PyPar doesn’t need manual annotation and is universally applicable. It first drives a data-dependence analysis to determine whether two pieces of code can run concurrently. The key is the use of a graph-theoretic approach. Next, it adopts a dynamic selection strategy to eliminate inefficient parallelisms. Finally, PyPar produces a parallelism report as well as a referential parallelized program, which is built by PyPar using one of the three parallelization methods (thread-based, processbased, and Ray-based). We have implemented a prototype of PyPar and evaluated it on six well-designed widely-used real-world Python packages: Scikit-Image, SciPy, librosa, trimesh, Scikit-learn and seaborn. In total, 1,240 functions are tested, and PyPar found 127 parallelizable functions among them. Based on manual filtering, only 7 of them are false positives (i.e., a 94.5% precision). The remaining 120 are parallelizable (almost 10% among all functions under test), and most of them can be efficiently sped up by gaining an acceleration of up to 90% , with an average of 44%. The acceleration in practice is close to theoretical estimation. The results show that even well-designed practical Python programs can be further parallelized for speeding up, and PyPar can bring effective and efficient parallelization on real-world Python programs.