193 citations · 457 across the 21 of their papers we have counts for
13 papers · 1 filter
Positive Label Is All You Need for Multi-Label Classification
Zhixiang Yuan, Kaixin Zhang, Tao Huang
Multi-label classification (MLC) faces challenges from label noise in training data due to annotating diverse semantic labels for each image. Current methods mainly target identify…
Knowledge Diffusion for Distillation
Tao Huang, Yuan Zhang, Mingkai Zheng +4
The representation gap between teacher and student is an emerging topic in knowledge distillation (KD). To reduce the gap and improve the performance, current methods often resort…
Avatar Knowledge Distillation: Self-ensemble Teacher Paradigm with Uncertainty
Yuan Zhang, Weihua Chen, Yichen Lu +3
Knowledge distillation is an effective paradigm for boosting the performance of pocket-size model, especially when multiple teacher models are available, the student would break th…
LightViT: Towards Light-Weight Convolution-Free Vision Transformers
Tao Huang, Lang Huang, Shan You +3
Vision transformers (ViTs) are usually considered to be less light-weight than convolutional neural networks (CNNs) due to the lack of inductive bias. Recent works thus resort to c…
Masked Distillation with Receptive Tokens
Tao Huang, Yuan Zhang, Shan You +4
Distilling from the feature maps can be fairly effective for dense prediction tasks since both the feature discriminability and localization priors can be well transferred. However…
Knowledge Distillation from A Stronger Teacher
Tao Huang, Shan You, Fei Wang +2
Unlike existing knowledge distillation methods focus on the baseline settings, where the teacher models and training strategies are not that strong and competing as state-of-the-ar…
DyRep: Bootstrapping Training with Dynamic Re-parameterization
Tao Huang, Shan You, Bohan Zhang +4
Structural re-parameterization (Rep) methods achieve noticeable improvements on simple VGG-style networks. Despite the prevalence, current Rep methods simply re-parameterize all op…
GreedyNASv2: Greedier Search with a Greedy Path Filter
Tao Huang, Shan You, Fei Wang +4
Training a good supernet in one-shot NAS methods is difficult since the search space is usually considerably huge (e.g., ). In order to enhance the supernet's evaluation a…
Prioritized Architecture Sampling with Monto-Carlo Tree Search
Xiu Su, Tao Huang, Yanxi Li +5
One-shot neural architecture search (NAS) methods significantly reduce the search cost by considering the whole search space as one network, which only needs to be trained once. Ho…
Locally Free Weight Sharing for Network Width Search
Xiu Su, Shan You, Tao Huang +4
Searching for network width is an effective way to slim deep neural networks with hardware budgets. With this aim, a one-shot supernet is usually leveraged as a performance evaluat…
Stretchable Cells Help DARTS Search Better
Tao Huang, Shan You, Yibo Yang +4
Differentiable neural architecture search (DARTS) has gained much success in discovering flexible and diverse cell types. To reduce the evaluation gap, the supernet is expected to…
Data Agnostic Filter Gating for Efficient Deep Networks
Xiu Su, Shan You, Tao Huang +5
To deploy a well-trained CNN model on low-end computation edge devices, it is usually supposed to compress or prune the model under certain computation budget (e.g., FLOPs). Curren…
GreedyNAS: Towards Fast One-Shot NAS with Greedy Supernet
Shan You, Tao Huang, Mingmin Yang +3
Training a supernet matters for one-shot neural architecture search (NAS) methods since it serves as a basic performance estimator for different architectures (paths). Current meth…