Author name cluster

Bing Su

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.

8 papers

1 author row

AAAI Conference 2025 Conference Paper

A Plug-and-Play Bregman ADMM Module for Inferring Event Branches in Temporal Point Processes

Qingmei Wang
Yuxin Wu
Yujie Long
Jing Huang
Fengyuan Ran
Bing Su
Hongteng Xu

An event sequence generated by a temporal point process is often associated with a hidden and structured event branching process that captures the triggering relations between its historical and current events. In this study, we design a new plug-and-play module based on the Bregman ADMM (BADMM) algorithm, which infers event branches associated with event sequences in the maximum likelihood estimation framework of temporal point processes (TPPs). Specifically, we formulate the inference of event branches as an optimization problem of event transition matrix under sparse and low-rank constraints, which is embedded in existing TPP models or their learning paradigms. We can implement this optimization problem based on subspace clustering and sparse group-lasso, respectively, and solve it using the Bregman ADMM algorithm, whose unrolling leads to the proposed BADMM module. When learning a classic TPP (e.g., Hawkes process) by the expectation-maximization algorithm, the BADMM module helps derive structured responsibility matrices in the E-step. Similarly, the BADMM module helps derive low-rank and sparse attention maps for the neural TPPs with self-attention layers. The structured responsibility matrices and attention maps, which work as learned event transition matrices, indicate event branches, e.g., inferring isolated events and those key events triggering many subsequent events. Experiments on both synthetic and real-world data show that plugging our BADMM module into existing TPP models and learning paradigms can improve model performance and provide us with interpretable structured event branches.

PDF Details DOI

AAAI Conference 2023 Conference Paper

Self-Supervised Action Representation Learning from Partial Spatio-Temporal Skeleton Sequences

Yujie Zhou
Haodong Duan
Anyi Rao
Bing Su
Jiaqi Wang

Self-supervised learning has demonstrated remarkable capability in representation learning for skeleton-based action recognition. Existing methods mainly focus on applying global data augmentation to generate different views of the skeleton sequence for contrastive learning. However, due to the rich action clues in the skeleton sequences, existing methods may only take a global perspective to learn to discriminate different skeletons without thoroughly leveraging the local relationship between different skeleton joints and video frames, which is essential for real-world applications. In this work, we propose a Partial Spatio-Temporal Learning (PSTL) framework to exploit the local relationship from a partial skeleton sequences built by a unique spatio-temporal masking strategy. Specifically, we construct a negative-sample-free triplet steam structure that is composed of an anchor stream without any masking, a spatial masking stream with Central Spatial Masking (CSM), and a temporal masking stream with Motion Attention Temporal Masking (MATM). The feature cross-correlation matrix is measured between the anchor stream and the other two masking streams, respectively. (1) Central Spatial Masking discards selected joints from the feature calculation process, where the joints with a higher degree of centrality have a higher possibility of being selected. (2) Motion Attention Temporal Masking leverages the motion of action and remove frames that move faster with a higher possibility. Our method achieves state-of-the-art performance on NTURGB+D 60, NTURGB+D 120 and PKU-MMD under various downstream tasks. Furthermore, to simulate the real-world scenarios, a practical evaluation is performed where some skeleton joints are lost in downstream tasks.In contrast to previous methods that suffer from large performance drops, our PSTL can still achieve remarkable results under this challenging setting, validating the robustness of our method.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

Log-Polar Space Convolution Layers

Bing Su
Ji-Rong Wen

Convolutional neural networks use regular quadrilateral convolution kernels to extract features. Since the number of parameters increases quadratically with the size of the convolution kernel, many popular models use small convolution kernels, resulting in small local receptive fields in lower layers. This paper proposes a novel log-polar space convolution (LPSC) layer, where the convolution kernel is elliptical and adaptively divides its local receptive field into different regions according to the relative directions and logarithmic distances. The local receptive field grows exponentially with the number of distance levels. Therefore, the proposed LPSC not only naturally encodes local spatial structures, but also greatly increases the single-layer receptive field while maintaining the number of parameters. We show that LPSC can be implemented with conventional convolution via log-polar space pooling and can be applied in any network architecture to replace conventional convolutions. Experiments on different tasks and datasets demonstrate the effectiveness of the proposed LPSC.

PDF Details

NeurIPS Conference 2022 Conference Paper

MetaMask: Revisiting Dimensional Confounder for Self-Supervised Learning

Jiangmeng Li
Wenwen Qiang
Yanan Zhang
Wenyi Mo
Changwen Zheng
Bing Su
Hui Xiong

As a successful approach to self-supervised learning, contrastive learning aims to learn invariant information shared among distortions of the input sample. While contrastive learning has yielded continuous advancements in sampling strategy and architecture design, it still remains two persistent defects: the interference of task-irrelevant information and sample inefficiency, which are related to the recurring existence of trivial constant solutions. From the perspective of dimensional analysis, we find out that the dimensional redundancy and dimensional confounder are the intrinsic issues behind the phenomena, and provide experimental evidence to support our viewpoint. We further propose a simple yet effective approach MetaMask, short for the dimensional Mask learned by Meta-learning, to learn representations against dimensional redundancy and confounder. MetaMask adopts the redundancy-reduction technique to tackle the dimensional redundancy issue and innovatively introduces a dimensional mask to reduce the gradient effects of specific dimensions containing the confounder, which is trained by employing a meta-learning paradigm with the objective of improving the performance of masked representations on a typical self-supervised task. We provide solid theoretical analyses to prove MetaMask can obtain tighter risk bounds for downstream classification compared to typical contrastive methods. Empirically, our method achieves state-of-the-art performance on various benchmarks.

PDF Details

NeurIPS Conference 2022 Conference Paper

SemMAE: Semantic-Guided Masking for Learning Masked Autoencoders

Gang Li
Heliang Zheng
Daqing Liu
Chaoyue Wang
Bing Su
Changwen Zheng

Recently, significant progress has been made in masked image modeling to catch up to masked language modeling. However, unlike words in NLP, the lack of semantic decomposition of images still makes masked autoencoding (MAE) different between vision and language. In this paper, we explore a potential visual analogue of words, i. e. , semantic parts, and we integrate semantic information into the training process of MAE by proposing a Semantic-Guided Masking strategy. Compared to widely adopted random masking, our masking strategy can gradually guide the network to learn various information, i. e. , from intra-part patterns to inter-part relations. In particular, we achieve this in two steps. 1) Semantic part learning: we design a self-supervised part learning method to obtain semantic parts by leveraging and refining the multi-head attention of a ViT-based encoder. 2) Semantic-guided MAE (SemMAE) training: we design a masking strategy that varies from masking a portion of patches in each part to masking a portion of (whole) parts in an image. Extensive experiments on various vision tasks show that SemMAE can learn better image representation by integrating semantic information. In particular, SemMAE achieves 84. 5% fine-tuning accuracy on ImageNet-1k, which outperforms the vanilla MAE by 1. 4%. In the semantic segmentation and fine-grained recognition tasks, SemMAE also brings significant improvements and yields the state-of-the-art performance.

PDF Details

TCS Journal 2020 Journal Article

Approximation algorithms for the three-machine proportionate mixed shop scheduling

Longcheng Liu
Yong Chen
Jianming Dong
Randy Goebel
Guohui Lin
Yue Luo
Guanqun Ni
Bing Su

A mixed shop is a manufacturing infrastructure designed to process a mixture of a set of flow-shop jobs and a set of open-shop jobs. Mixed shops are in general much more complex to schedule than flow-shops and open-shops, and have been studied since the 1980's. We consider the three machine proportionate mixed shop problem denoted as M 3 | p r p t | C max, in which by “proportionate” each job has equal processing times on all three machines. Koulamas and Kyparisis (2015) [6] showed that the problem is solvable in polynomial time in some very special cases; for the non-solvable case, they proposed a 5/3-approximation algorithm. In this paper, we first present an improved 4/3-approximation algorithm and show that this ratio of 4/3 is asymptotically tight; when the largest job is a flow-shop job, we then present a fully polynomial-time approximation scheme (FPTAS). On the negative side, while the F 3 | p r p t | C max problem is polynomial-time solvable, we show an interesting hardness result that adding one open-shop job to the job set makes the problem NP-hard if this open-shop job is larger than any flow-shop job. We are able to design an FPTAS for this special case too.

Details DOI

TCS Journal 2020 Journal Article

Open-shop scheduling for unit jobs under precedence constraints

Yong Chen
Randy Goebel
Guohui Lin
Bing Su
An Zhang

We study open-shop scheduling for unit jobs under precedence constraints, where if one job precedes another job then it has to be finished before the other job can start to be processed. For the three-machine open-shop to minimize the makespan, we first present a simple 5/3-approximation algorithm based on a partition of the job set into agreeable layers using the natural layered representation of the precedence graph, which is directed acyclic. We then show a greedy algorithm to reduce the number of singleton-job layers, resulting in an improved partition, which leads to a 4/3-approximation algorithm. Both approximation algorithms apply to the general m-machine open-shops too.

Details DOI

TCS Journal 2015 Journal Article

Minimax regret 1-sink location problem in dynamic path networks

Yuya Higashikawa
John Augustine
Siu-Wing Cheng
Mordecai J. Golin
Naoki Katoh
Guanqun Ni
Bing Su
Yinfeng Xu

This paper considers the minimax regret 1-sink location problem in dynamic path networks. In our model, a dynamic path network consists of an undirected path with positive edge lengths and uniform edge capacity, and each vertex supply which is nonnegative value is unknown but only the interval of supply is known. A particular assignment of supply to each vertex is called a scenario. Under any scenario, the cost of a sink location is defined as the minimum time to complete the evacuation for all supplies (evacuees), and the regret of a sink location x is defined as the cost of x minus the cost of the optimal sink location. Then, the problem is to find a point as a sink such that the maximum regret for all possible scenarios is minimized. We propose an O ( n log ⁡ n ) time algorithm for the minimax regret 1-sink location problem in dynamic path networks with uniform capacity, where n is the number of vertices in the network.

Details DOI