no code implementations • 3 Apr 2024 • Haofeng Yuan, Rongping Zhu, Wanlu Yang, Shiji Song, Keyou You, Yuli Zhang
The traveling purchaser problem (TPP) is an important combinatorial optimization problem with broad applications.
1 code implementation • 11 Mar 2024 • Chaoqun Du, Yulin Wang, Shiji Song, Gao Huang
To overcome this obstacle, we propose a novel probabilistic contrastive (ProCo) learning algorithm that estimates the data distribution of the samples from each class in the feature space, and samples contrastive pairs accordingly.
Ranked #8 on Long-tail Learning on iNaturalist 2018
no code implementations • 19 Feb 2024 • Qisen Yang, Zekun Wang, Honghui Chen, Shenzhi Wang, Yifan Pu, Xin Gao, Wenhao Huang, Shiji Song, Gao Huang
Psychological measurement is essential for mental health, self-understanding, and personal development.
no code implementations • 28 Dec 2023 • Rui Huang, Songyou Peng, Ayca Takmaz, Federico Tombari, Marc Pollefeys, Shiji Song, Gao Huang, Francis Engelmann
Therefore, we explore the use of image segmentation foundation models to automatically generate training labels for 3D segmentation.
no code implementations • 21 Dec 2023 • Haofeng Yuan, Lichang Fang, Shiji Song
Column generation (CG) is one of the most successful approaches for solving large-scale linear programming (LP) problems.
no code implementations • 15 Dec 2023 • Zhuofan Xia, Dongchen Han, Yizeng Han, Xuran Pan, Shiji Song, Gao Huang
Generalized Referring Expression Segmentation (GRES) extends the scope of classic RES to refer to multiple objects in one expression or identify the empty targets absent in the image.
Generalized Referring Expression Segmentation Referring Expression +1
2 code implementations • 14 Dec 2023 • Dongchen Han, Tianzhu Ye, Yizeng Han, Zhuofan Xia, Shiji Song, Gao Huang
Specifically, the Agent Attention, denoted as a quadruple $(Q, A, K, V)$, introduces an additional set of agent tokens $A$ into the conventional attention module.
1 code implementation • 7 Dec 2023 • Jiayi Guo, Xingqian Xu, Yifan Pu, Zanlin Ni, Chaofei Wang, Manushree Vasu, Shiji Song, Gao Huang, Humphrey Shi
Specifically, we introduce Step-wise Variation Regularization to enforce the proportion between the variations of an arbitrary input latent and that of the output image is a constant at any diffusion training step.
1 code implementation • NeurIPS 2023 • Shenzhi Wang, Qisen Yang, Jiawei Gao, Matthieu Gaetan Lin, Hao Chen, Liwei Wu, Ning Jia, Shiji Song, Gao Huang
Existing solutions tackle this problem by imposing a policy constraint on the policy improvement objective in both offline and online learning.
2 code implementations • NeurIPS 2023 • Yang Yue, Rui Lu, Bingyi Kang, Shiji Song, Gao Huang
We first identify a fundamental pattern, self-excitation, as the primary cause of Q-value estimation divergence in offline RL.
no code implementations • 2 Oct 2023 • Shenzhi Wang, Chang Liu, Zilong Zheng, Siyuan Qi, Shuo Chen, Qisen Yang, Andrew Zhao, Chaofei Wang, Shiji Song, Gao Huang
This study utilizes the intricate Avalon game as a testbed to explore LLMs' potential in deceptive environments.
no code implementations • 4 Sep 2023 • Qisen Yang, Shenzhi Wang, Qihang Zhang, Gao Huang, Shiji Song
Offline reinforcement learning (RL) optimizes the policy on a previously collected dataset without any interactions with the environment, yet usually suffers from the distributional shift problem.
1 code implementation • 4 Sep 2023 • Zhuofan Xia, Xuran Pan, Shiji Song, Li Erran Li, Gao Huang
On the one hand, using dense attention in ViT leads to excessive memory and computational cost, and features can be influenced by irrelevant parts that are beyond the region of interests.
Ranked #4 on Object Detection on COCO 2017
1 code implementation • 30 Aug 2023 • Yizeng Han, Zeyu Liu, Zhihang Yuan, Yifan Pu, Chaofei Wang, Shiji Song, Gao Huang
Dynamic computation has emerged as a promising avenue to enhance the inference efficiency of deep networks.
no code implementations • 27 Aug 2023 • Yulin Wang, Yizeng Han, Chaofei Wang, Shiji Song, Qi Tian, Gao Huang
Over the past decade, deep learning models have exhibited considerable advancements, reaching or even exceeding human-level performance in a range of visual perception tasks.
1 code implementation • 8 Aug 2023 • Honghui Wang, Lu Lu, Shiji Song, Gao Huang
To avoid the inefficient manual selection and to alleviate the optimization difficulty of PINNs, we introduce adaptive activation functions to search for the optimal function when solving different problems.
1 code implementation • ICCV 2023 • Dongchen Han, Xuran Pan, Yizeng Han, Shiji Song, Gao Huang
The quadratic computation complexity of self-attention has been a persistent challenge when applying Transformer models to vision tasks.
1 code implementation • ICCV 2023 • Yizeng Han, Dongchen Han, Zeyu Liu, Yulin Wang, Xuran Pan, Yifan Pu, Chao Deng, Junlan Feng, Shiji Song, Gao Huang
Early exits are placed exclusively within the classification branch, thus eliminating the need for linear separability in low-level features.
2 code implementations • 8 Jun 2023 • Yang Yue, Bingyi Kang, Xiao Ma, Qisen Yang, Gao Huang, Shiji Song, Shuicheng Yan
OPER is a plug-and-play component for offline RL algorithms.
no code implementations • 6 Jun 2023 • Qisen Yang, Shenzhi Wang, Matthieu Gaetan Lin, Shiji Song, Gao Huang
In particular, online fine-tuning has become a commonly used method to correct the erroneous estimates of out-of-distribution data learned in the offline training phase.
1 code implementation • CVPR 2023 • Xuran Pan, Tianzhu Ye, Zhuofan Xia, Shiji Song, Gao Huang
Self-attention mechanism has been a key factor in the recent progress of Vision Transformer (ViT), which enables adaptive feature extraction from global contexts.
1 code implementation • CVPR 2023 • Jiayi Guo, Chaofei Wang, You Wu, Eric Zhang, Kai Wang, Xingqian Xu, Shiji Song, Humphrey Shi, Gao Huang
Recently, CLIP-guided image synthesis has shown appealing performance on adapting a pre-trained source-domain generator to an unseen target domain.
1 code implementation • ICCV 2023 • Yifan Pu, Yiru Wang, Zhuofan Xia, Yizeng Han, Yulin Wang, Weihao Gan, Zidong Wang, Shiji Song, Gao Huang
In our ARC module, the convolution kernels rotate adaptively to extract object features with varying orientations in different images, and an efficient conditional computation mechanism is introduced to accommodate the large orientation variations of objects within an image.
Ranked #3 on Object Detection In Aerial Images on DOTA (using extra training data)
no code implementations • 18 Jan 2023 • Rui Huang, Xuran Pan, Henry Zheng, Haojun Jiang, Zhifeng Xie, Shiji Song, Gao Huang
During the pre-training stage, we establish the correspondence of images and point clouds based on the readily available RGB-D data and use contrastive learning to align the image and point cloud representations.
no code implementations • 21 Dec 2022 • Fei Dong, Xingchen Li, Keyou You, Shiji Song
This work studies the standoff tracking problem to drive an unmanned aerial vehicle (UAV) to slide on a desired circle over a moving target at a constant height.
1 code implementation • 17 Nov 2022 • Haojun Jiang, Jianke Zhang, Rui Huang, Chunjiang Ge, Zanlin Ni, Jiwen Lu, Jie zhou, Shiji Song, Gao Huang
However, as pre-trained models are scaling up, fully fine-tuning them on text-video retrieval datasets has a high risk of overfitting.
1 code implementation • ICCV 2023 • Yulin Wang, Yang Yue, Rui Lu, Tianjiao Liu, Zhao Zhong, Shiji Song, Gao Huang
It is also effective for self-supervised learning (e. g., MAE).
no code implementations • 17 Oct 2022 • Xuran Pan, Tianzhu Ye, Dongchen Han, Shiji Song, Gao Huang
Recent years have witnessed the fast development of large-scale pre-training frameworks that can extract multi-modal representations in a unified form and achieve promising performances when transferred to downstream tasks.
1 code implementation • 12 Oct 2022 • Chaofei Wang, Qisen Yang, Rui Huang, Shiji Song, Gao Huang
Knowledge distillation is an effective approach to learn compact models (students) with the supervision of large and strong models (teachers).
2 code implementations • 12 Oct 2022 • Yizeng Han, Zhihang Yuan, Yifan Pu, Chenhao Xue, Shiji Song, Guangyu Sun, Gao Huang
The latency prediction model can efficiently estimate the inference latency of dynamic networks by simultaneously considering algorithms, scheduling strategies, and hardware properties.
no code implementations • 27 Sep 2022 • Yulin Wang, Yang Yue, Xinhong Xu, Ali Hassani, Victor Kulikov, Nikita Orlov, Shiji Song, Humphrey Shi, Gao Huang
Recent research has revealed that reducing the temporal and spatial redundancy are both effective approaches towards efficient video recognition, e. g., allocating the majority of computation to a task-relevant subset of frames or the most valuable image regions of each frame.
1 code implementation • 18 Sep 2022 • Xuran Pan, Zihang Lai, Shiji Song, Gao Huang
In this paper, we present a novel learning framework, ActiveNeRF, aiming to model a 3D scene with a constrained input budget.
1 code implementation • 17 Sep 2022 • Yizeng Han, Yifan Pu, Zihang Lai, Chaofei Wang, Shiji Song, Junfen Cao, Wenhui Huang, Chao Deng, Gao Huang
Intuitively, easy samples, which generally exit early in the network during inference, should contribute more to training early classifiers.
no code implementations • 4 Jul 2022 • Haofeng Yuan, Peng Jiang, Shiji Song
In this paper, we propose an improved column generation algorithm with neural prediction (CG-P) for solving graph-based set covering problems.
1 code implementation • CVPR 2022 • Haojun Jiang, Yuanze Lin, Dongchen Han, Shiji Song, Gao Huang
Our method leverages an off-the-shelf object detector to identify visual objects from unlabeled images, and then language queries for these objects are obtained in an unsupervised fashion with a pseudo-query generation module.
no code implementations • 25 Feb 2022 • Chaofei Wang, Shaowei Zhang, Shiji Song, Gao Huang
We save a moderate number of intermediate models from the training process of the teacher model uniformly, and then integrate the knowledge of these intermediate models by ensemble technique.
1 code implementation • 14 Feb 2022 • Chunjiang Ge, Rui Huang, Mixue Xie, Zihang Lai, Shiji Song, Shuang Li, Gao Huang
Unsupervised domain adaption (UDA) aims to adapt models learned from a well-annotated source domain to a target domain, where only unlabeled samples are given.
1 code implementation • 9 Jan 2022 • Gao Huang, Yulin Wang, Kangchen Lv, Haojun Jiang, Wenhui Huang, Pengfei Qi, Shiji Song
Spatial redundancy widely exists in visual recognition tasks, i. e., discriminative features in an image or video frame usually correspond to only a subset of pixels, while the remaining regions are irrelevant to the task at hand.
2 code implementations • CVPR 2022 • Zhuofan Xia, Xuran Pan, Shiji Song, Li Erran Li, Gao Huang
On the one hand, using dense attention e. g., in ViT, leads to excessive memory and computational cost, and features can be influenced by irrelevant parts which are beyond the region of interests.
Ranked #107 on Object Detection on COCO test-dev
1 code implementation • CVPR 2022 • Chenxin Tao, Honghui Wang, Xizhou Zhu, Jiahua Dong, Shiji Song, Gao Huang, Jifeng Dai
These methods appear to be quite different in the designed loss functions from various motivations.
no code implementations • 6 Dec 2021 • Wenjie Shi, Gao Huang, Shiji Song, Cheng Wu
TSCI model builds on the formulation of temporal causality, which reflects the temporal causal relations between sequential observations and decisions of RL agent.
2 code implementations • CVPR 2022 • Xuran Pan, Chunjiang Ge, Rui Lu, Shiji Song, Guanfu Chen, Zeyi Huang, Gao Huang
In this paper, we show that there exists a strong underlying relation between them, in the sense that the bulk of computations of these two paradigms are in fact done with the same operation.
no code implementations • 13 Sep 2021 • Chaofei Wang, Shiji Song, Qisen Yang, Xiang Li, Gao Huang
As a data augmentation method, FOT can be conveniently applied to any existing few shot learning algorithm and greatly improve its performance on FG-FSL tasks.
no code implementations • ICCV 2021 • Chaofei Wang, Jiayu Xiao, Yizeng Han, Qisen Yang, Shiji Song, Gao Huang
The backbone of traditional CNN classifier is generally considered as a feature extractor, followed by a linear layer which performs the classification.
2 code implementations • NeurIPS 2021 • Yulin Wang, Rui Huang, Shiji Song, Zeyi Huang, Gao Huang
Inspired by this phenomenon, we propose a Dynamic Transformer to automatically configure a proper number of tokens for each input image.
Ranked #29 on Image Classification on CIFAR-100 (using extra training data)
1 code implementation • ICCV 2021 • Yulin Wang, Zhaoxi Chen, Haojun Jiang, Shiji Song, Yizeng Han, Gao Huang
In this paper, we explore the spatial redundancy in video recognition with the aim to improve the computational efficiency.
1 code implementation • CVPR 2021 • Le Yang, Haojun Jiang, Ruojin Cai, Yulin Wang, Shiji Song, Gao Huang, Qi Tian
Reusing features in deep networks through dense connectivity is an effective way to achieve high computational efficiency.
no code implementations • 9 Feb 2021 • Yizeng Han, Gao Huang, Shiji Song, Le Yang, Honghui Wang, Yulin Wang
Dynamic neural network is an emerging research topic in deep learning.
1 code implementation • 26 Jan 2021 • Yulin Wang, Zanlin Ni, Shiji Song, Le Yang, Gao Huang
Due to the need to store the intermediate activations for back-propagation, end-to-end (E2E) training of deep networks usually suffers from high GPUs memory footprint.
no code implementations • 1 Jan 2021 • Wenjie Shi, Tianchi Cai, Shiji Song, Lihong Gu, Jinjie Gu, Gao Huang
We theoretically show that AdaPT produces a tight upper bound on the distributional deviation between the learned policy and the behavior policy, and this upper bound is the minimum requirement to guarantee policy improvement at each iteration.
no code implementations • ICLR 2021 • Yulin Wang, Zanlin Ni, Shiji Song, Le Yang, Gao Huang
As InfoPro loss is difficult to compute in its original form, we derive a feasible upper bound as a surrogate optimization objective, yielding a simple but effective algorithm.
no code implementations • 1 Jan 2021 • Xuran Pan, Shiji Song, Gao Huang
In this paper, we take a step forward to establish a unified framework for convolution-based graph neural networks, by formulating the basic graph convolution operation as an optimization problem in the graph Fourier space.
1 code implementation • CVPR 2021 • Xuran Pan, Zhuofan Xia, Shiji Song, Li Erran Li, Gao Huang
In this paper, we propose Pointformer, a Transformer backbone designed for 3D point clouds to learn features effectively.
1 code implementation • NeurIPS 2020 • Yulin Wang, Kangchen Lv, Rui Huang, Shiji Song, Le Yang, Gao Huang
The accuracy of deep convolutional neural networks (CNNs) generally improves when fueled with high resolution images.
1 code implementation • 21 Jul 2020 • Yulin Wang, Gao Huang, Shiji Song, Xuran Pan, Yitong Xia, Cheng Wu
The proposed method is inspired by the intriguing property that deep networks are effective in learning linearized features, i. e., certain directions in the deep feature space correspond to meaningful semantic transformations, e. g., changing the background or view angle of an object.
no code implementations • 5 Jul 2020 • Yulin Wang, Jiayi Guo, Shiji Song, Gao Huang
In this paper, we propose a novel meta-learning based SSL algorithm (Meta-Semi) that requires tuning only one additional hyper-parameter, compared with a standard supervised deep learning algorithm, to achieve competitive performance under various conditions of SSL.
1 code implementation • 16 Mar 2020 • Wenjie Shi, Gao Huang, Shiji Song, Zhuoyuan Wang, Tingyu Lin, Cheng Wu
Deep reinforcement learning (RL) has recently led to many breakthroughs on a range of complex control tasks.
2 code implementations • CVPR 2020 • Le Yang, Yizeng Han, Xi Chen, Shiji Song, Jifeng Dai, Gao Huang
Adaptive inference is an effective mechanism to achieve a dynamic tradeoff between accuracy and computational cost in deep networks.
no code implementations • 6 Mar 2020 • Kaichen Zhou, Shiji Song, Gao Huang, Wu Cheng, Quan Zhou
Specifically, the proposed algorithm can be used to estimate the upper and lower bounds of the updated classifier's coefficient matrix with a low computational complexity related to the size of the updated dataset.
no code implementations • 6 Mar 2020 • Kaichen Zhou, Shiji Song, Anke Xue, Keyou You, Hui Wu
Then we develop two algorithms for optimizing the energy efficiency of train operation.
no code implementations • 27 Nov 2019 • Ya-Chu Hsu, Hui Wu, Keyou You, Shiji Song
This paper provides a selected review on RL based control for AUVs with the focus on applications of RL to low-level control tasks for underwater regulation and tracking.
Robotics
1 code implementation • NeurIPS 2019 • Yulin Wang, Xuran Pan, Shiji Song, Hong Zhang, Cheng Wu, Gao Huang
Our work is motivated by the intriguing property that deep networks are surprisingly good at linearizing features, such that certain directions in the deep feature space correspond to meaningful semantic transformations, e. g., adding sunglasses or changing backgrounds.
no code implementations • 7 Sep 2019 • Wenjie Shi, Shiji Song, Cheng Wu, C. L. Philip Chen
Different from existing policy gradient methods which employ single actor-critic but cannot realize satisfactory tracking control accuracy and stable learning, our proposed algorithm can achieve high-level tracking control accuracy of AUVs and stable learning by applying a hybrid actors-critics architecture, where multiple actors and critics are trained to learn a deterministic policy and action-value function, respectively.
no code implementations • 7 Sep 2019 • Wenjie Shi, Shiji Song, Cheng Wu
Then, we present an off-policy actor-critic, model-free maximum entropy deep RL algorithm called deep soft policy gradient (DSPG) by combining soft policy gradient with soft Bellman equation.
1 code implementation • NeurIPS 2019 • Wenjie Shi, Shiji Song, Hui Wu, Ya-Chu Hsu, Cheng Wu, Gao Huang
To tackle this problem, we propose a general acceleration method for model-free, off-policy deep RL algorithms by drawing the idea underlying regularized Anderson acceleration (RAA), which is an effective approach to accelerating the solving of fixed point problems with perturbations.
no code implementations • 6 Sep 2014 • Quan Zhou, Wenlin Chen, Shiji Song, Jacob R. Gardner, Kilian Q. Weinberger, Yixin Chen
Our second contribution is to derive a practical algorithm based on this reduction.