no code implementations • 27 Oct 2023 • Zeren Zhang, Ran Chen, Jinwen Ma
This paper introduces an approach to enhance seismic fault recognition through self-supervised pretraining.
1 code implementation • 4 Mar 2023 • Zhijian Zhuo, Yifei Wang, Jinwen Ma, Yisen Wang
In this work, we propose a unified theoretical understanding for existing variants of non-contrastive learning.
1 code implementation • 7 Aug 2022 • Zhengyang Shen, Tao Hong, Qi She, Jinwen Ma, Zhouchen Lin
Steerable models can provide very general and flexible equivariance by formulating equivariance requirements in the language of representation theory and feature fields, which has been recognized to be effective for many vision tasks.
no code implementations • 10 Oct 2021 • Yuan-Ching Lin, Jinwen Ma
Representing a text as a graph for obtaining automatic text summarization has been investigated for over ten years.
no code implementations • 13 Sep 2021 • Daqing Wu, Xiao Luo, Zeyu Ma, Chong Chen, Minghua Deng, Jinwen Ma
Nowadays, E-commerce is increasingly integrated into our daily lives.
1 code implementation • 25 May 2021 • Xiao Luo, Daqing Wu, Yiyang Gu, Chong Chen, Luchen Liu, Jinwen Ma, Ming Zhang, Minghua Deng, Jianqiang Huang, Xian-Sheng Hua
Besides, CHCF integrates criterion learning and user preference learning into a unified framework, which can be trained jointly for the interaction prediction of the target behavior.
no code implementations • 13 May 2021 • Xiao Luo, Zeyu Ma, Daqing Wu, Huasong Zhong, Chong Chen, Jinwen Ma, Minghua Deng
Hashing has been widely used in approximate nearest neighbor search for its storage and computational efficiency.
no code implementations • 8 Apr 2021 • Zhengyang Shen, Tiancheng Shen, Zhouchen Lin, Jinwen Ma
Spherical signals exist in many applications, e. g., planetary data, LiDAR scans and digitalization of 3D objects, calling for models that can process spherical data effectively.
1 code implementation • NeurIPS 2020 • Wenpeng Hu, Mengyu Wang, Qi Qin, Jinwen Ma, Bing Liu
Existing neural network based one-class learning methods mainly use various forms of auto-encoders or GAN style adversarial training to learn a latent representation of the given one class of data.
no code implementations • COLING 2020 • Wenpeng Hu, Ran Le, Bing Liu, Jinwen Ma, Dongyan Zhao, Rui Yan
Understanding neural models is a major topic of interest in the deep learning community.
no code implementations • 15 Oct 2020 • Xiao Luo, Daqing Wu, Zeyu Ma, Chong Chen, Minghua Deng, Jinwen Ma, Zhongming Jin, Jianqiang Huang, Xian-Sheng Hua
However, due to the inefficient representation ability of the pre-trained model, many false positives and negatives in local semantic similarity will be introduced and lead to error propagation during the hash code learning.
3 code implementations • ICML 2020 • Zhengyang Shen, Lingshen He, Zhouchen Lin, Jinwen Ma
In implementation, we discretize the system using the numerical schemes of PDOs, deriving approximately equivariant convolutions (PDO-eConvs).
Ranked #1 on Image Classification on MNIST-rot-12
no code implementations • 16 Jun 2020 • Jie An, Tao Li, Hao-Zhi Huang, Li Shen, Xuan Wang, Yongyi Tang, Jinwen Ma, Wei Liu, Jiebo Luo
Extracting effective deep features to represent content and style information is the key to universal style transfer.
1 code implementation • 16 Jun 2020 • Ao Zhang, Jinwen Ma
Graph neural networks (GNNs) achieve remarkable performance for tasks on graph data.
2 code implementations • 21 Nov 2019 • Ya Wang, Dongliang He, Fu Li, Xiang Long, Zhichao Zhou, Jinwen Ma, Shilei Wen
In this paper, we propose a label graph superimposing framework to improve the conventional GCN+CNN framework developed for multi-label recognition in the following two aspects.
Ranked #28 on Multi-Label Classification on MS-COCO
1 code implementation • COLING 2020 • Wenpeng Hu, Mengyu Wang, Bing Liu, Feng Ji, Haiqing Chen, Dongyan Zhao, Jinwen Ma, Rui Yan
The key idea of the proposed approach is to use a Forward Transformation to transform dense representations to sparse representations.
no code implementations • 25 Sep 2019 • Wenpeng Hu, Ran Le, Bing Liu, Feng Ji, Haiqing Chen, Dongyan Zhao, Jinwen Ma, Rui Yan
Positive-unlabeled (PU) learning learns a binary classifier using only positive and unlabeled examples without labeled negative examples.
no code implementations • 6 Jul 2019 • Jie An, Haoyi Xiong, Jiebo Luo, Jun Huan, Jinwen Ma
Given a pair of images as the source of content and the reference of style, existing solutions usually first train an auto-encoder (AE) to reconstruct the image using deep features and then embeds pre-defined style transfer modules into the AE reconstruction procedure to transfer the style of the reconstructed image through modifying the deep features.
no code implementations • 6 Jun 2019 • Jie An, Haoyi Xiong, Jinwen Ma, Jiebo Luo, Jun Huan
Finally compared to existing universal style transfer networks for photorealistic rendering such as PhotoWCT that stacks multiple well-trained auto-encoders and WCT transforms in a non-end-to-end manner, the architectures designed by StyleNAS produce better style-transferred images with details preserving, using a tiny number of operators/parameters, and enjoying around 500x inference time speed-up.
1 code implementation • 31 May 2019 • Wenpeng Hu, Zhangming Chan, Bing Liu, Dongyan Zhao, Jinwen Ma, Rui Yan
Existing neural models for dialogue response generation assume that utterances are sequentially organized.
no code implementations • ICLR 2019 • Wenpeng Hu, Zhou Lin, Bing Liu, Chongyang Tao, Zhengwei Tao, Jinwen Ma, Dongyan Zhao, Rui Yan
Several continual learning methods have been proposed to address the problem.
no code implementations • ICLR 2019 • Zhanxing Zhu, Jingfeng Wu, Bing Yu, Lei Wu, Jinwen Ma
Along this line, we theoretically study a general form of gradient based optimization dynamics with unbiased noise, which unifies SGD and standard Langevin dynamics.
no code implementations • ICLR 2019 • Wenpeng Hu, Zhengwei Tao, Zhanxing Zhu, Bing Liu, Zhou Lin, Jinwen Ma, Dongyan Zhao, Rui Yan
A large amount of parallel data is needed to train a strong neural machine translation (NMT) system.
no code implementations • 23 Apr 2019 • Tao Li, Jinwen Ma
Tensor completion and robust principal component analysis have been widely used in machine learning while the key problem relies on the minimization of a tensor rank that is very challenging.
no code implementations • 11 Mar 2019 • Yixin Li, Shengqin Tang, Yun Ye, Jinwen Ma
Fashion landmark detection is a challenging task even using the current deep learning techniques, due to the large variation and non-rigid deformation of clothes.
1 code implementation • CVPR 2019 • Bing Yu, Jingfeng Wu, Jinwen Ma, Zhanxing Zhu
The proposed TNAR is composed by two complementary parts, the tangent adversarial regularization (TAR) and the normal adversarial regularization (NAR).
2 code implementations • ECCV 2018 • Taihong Xiao, Jiapeng Hong, Jinwen Ma
Recent studies on face attribute transfer have achieved great success.
1 code implementation • ICLR 2019 • Zhanxing Zhu, Jingfeng Wu, Bing Yu, Lei Wu, Jinwen Ma
Along this line, we study a general form of gradient based optimization dynamics with unbiased noise, which unifies SGD and standard Langevin dynamics.
no code implementations • ICLR 2018 • Wenpeng Hu, Bing Liu, Rui Yan, Dongyan Zhao, Jinwen Ma
In the paper, we propose a new question generation problem, which also requires the input of a target topic in addition to a piece of descriptive text.
1 code implementation • ICLR 2018 • Taihong Xiao, Jiapeng Hong, Jinwen Ma
Disentangling factors of variation has become a very challenging problem on representation learning.
no code implementations • CVPR 2015 • Yunsheng Jiang, Jinwen Ma
This paper presents effective combination models with certain combination features for human detection.