Implicit Semantic Data Augmentation for Deep Networks
- Yulin Wang, Xuran Pan, Shiji Song, Hong Zhang, Cheng Wu, Gao Huang
- Computer ScienceNeural Information Processing Systems
- 26 September 2019
This work shows that the proposed ISDA amounts to minimizing a novel robust CE loss, which adds negligible extra computational cost to a normal training procedure, and consistently improves the generalization performance of popular deep models (ResNets and DenseNets) on a variety of datasets.
Regularizing Deep Networks With Semantic Data Augmentation
- Yulin Wang, Gao Huang, Shiji Song, Xuran Pan, Yitong Xia, Cheng Wu
- Computer ScienceIEEE Transactions on Pattern Analysis and Machineā¦
- 21 July 2020
The proposed implicit semantic data augmentation (ISDA) algorithm amounts to minimizing a novel robust CE loss, which adds minimal extra computational cost to a normal training procedure, and can be applied to semi-supervised learning tasks under the consistency regularization framework.
Adaptive Focus for Efficient Video Recognition
- Yulin Wang, Zhaoxi Chen, Haojun Jiang, Shiji Song, Yizeng Han, Gao Huang
- Computer ScienceIEEE International Conference on Computer Vision
- 7 May 2021
This paper model the patch localization problem as a sequential decision task, and proposes a reinforcement learning based approach for efficient spatially adaptive video recognition (AdaFocus), whose features are used by a recurrent policy network to localize the most task-relevant regions.
Revisiting Locally Supervised Learning: an Alternative to End-to-end Training
- Yulin Wang, Z. Ni, Shiji Song, Le Yang, Gao Huang
- Computer ScienceInternational Conference on Learningā¦
- 26 January 2021
An information propagation (InfoPro) loss is proposed, which encourages local modules to preserve as much useful information as possible, while progressively discard task-irrelevant information, and is capable of achieving competitive performance with less than 40% memory footprint compared to E2E training.
Dynamic Neural Networks: A Survey
- Yizeng Han, Gao Huang, Shiji Song, Le Yang, Honghui Wang, Yulin Wang
- Computer ScienceIEEE Transactions on Pattern Analysis and Machineā¦
- 9 February 2021
This survey comprehensively review this rapidly developing area of dynamic networks by dividing dynamic networks into three main categories: sample-wise dynamic models that process each sample with data-dependent architectures or parameters; spatial-wiseynamic networks that conduct adaptive computation with respect to different spatial locations of image data; and temporal-wise Dynamic networks that perform adaptive inference along the temporal dimension for sequential data.
Glance and Focus: a Dynamic Approach to Reducing Spatial Redundancy in Image Classification
- Yulin Wang, Kangchen Lv, Rui Huang, Shiji Song, Le Yang, Gao Huang
- Computer ScienceNeural Information Processing Systems
- 11 October 2020
This work proposes a novel framework that performs efficient image classification by processing a sequence of relatively small inputs, which are strategically selected from the original image with reinforcement learning, which consistently improves the computational efficiency of a wide variety of deep models.
MetaSAug: Meta Semantic Augmentation for Long-Tailed Visual Recognition
- Shuang Li, Kaixiong Gong, Chi Harold Liu, Yulin Wang, Feng Qiao, Xinjing Cheng
- Computer ScienceComputer Vision and Pattern Recognition
- 23 March 2021
This paper addresses the issue of imbalance in real-world training data by augmenting minority classes with a recently proposed implicit semantic data augmentation (ISDA) algorithm, which produces diversified augmented samples by translating deep features along many semantically meaningful directions.
Not All Images are Worth 16x16 Words: Dynamic Vision Transformers with Adaptive Sequence Length
- Yulin Wang, Rui Huang, Shiji Song, Zeyi Huang, Gao Huang
- Computer ScienceArXiv
- 2021
This paper argues that every image has its own characteristics, and ideally the token number should be conditioned on each individual input, and proposes a Dynamic Transformer to automatically configure a proper number of tokens for each input image.
Not All Images are Worth 16x16 Words: Dynamic Transformers for Efficient Image Recognition
- Yulin Wang, Rui Huang, S. Song, Zeyi Huang, Gao Huang
- Computer ScienceNeural Information Processing Systems
- 31 May 2021
This paper argues that every image has its own characteristics, and ideally the token number should be conditioned on each individual input, and proposes a Dynamic Transformer to automatically configure a proper number of tokens for each input image.
Transferable Semantic Augmentation for Domain Adaptation
- Shuang Li, Mixue Xie, Kaixiong Gong, Chi Harold Liu, Yulin Wang, Wei Li
- Computer ScienceComputer Vision and Pattern Recognition
- 23 March 2021
This work proposes a Transferable Semantic Augmentation approach to enhance the classifier adaptation ability through implicitly generating source features towards target semantics, inspired by the fact that deep feature transformation towards a certain direction can be represented as meaningful semantic altering in the original input space.
...
...