paper-with-me

홈 › Papers

Slimmable Domain Adaptation

2022-06-14 · CVPR 2022 1 · Rang Meng, WeiJie Chen, Shicai Yang, Jie Song, Luojun Lin, Di Xie, ShiLiang Pu, Xinchao Wang, Mingli Song, Yueting Zhuang

Vanilla unsupervised domain adaptation methods tend to optimize the model with fixed neural architecture, which is not very practical in real-world scenarios since the target data is usually processed by different resource-limited devices. It is therefore of great necessity to facilitate architecture adaptation across various devices. In this paper, we introduce a simple framework, Slimmable Domain Adaptation, to improve cross-domain generalization with a weight-sharing model bank, from which models of different capacities can be sampled to accommodate different accuracy-efficiency trade-offs. The main challenge in this framework lies in simultaneously boosting the adaptation performance of numerous models in the model bank. To tackle this problem, we develop a Stochastic EnsEmble Distillation method to fully exploit the complementary knowledge in the model bank for inter-model interaction. Nevertheless, considering the optimization conflict between inter-model interaction and intra-model adaptation, we augment the existing bi-classifier domain confusion architecture into an Optimization-Separated Tri-Classifier counterpart. After optimizing the model bank, architecture adaptation is leveraged via our proposed Unsupervised Performance Evaluation Metric. Under various resource constraints, our framework surpasses other competing approaches by a very large margin on multiple benchmarks. It is also worth emphasizing that our framework can preserve the performance improvement against the source-only model even when the computing complexity is reduced to $1/64$. Code will be available at https://github.com/hikvision-research/SlimDA.

📄 PDF Abstract BibTeX arXiv:2206.06620

Code (1)

hik-lab/slimda 공식 구현

Tasks

Domain AdaptationDomain GeneralizationUnsupervised Domain Adaptation

Similar Papers 제목 키워드 기반

Three Guidelines You Should Know for Universally Slimmable Self-Supervised Learning

2023-03-13 · CVPR 2023 1 · Yun-Hao Cao, Peiqin Sun, Shuchang Zhou

We propose universally slimmable self-supervised learning (dubbed as US3L) to achieve better accuracy-efficiency trade-offs for deploying self-supervised models across different devices. We observe that direct adaptation…

Instance Segmentationobject-detectionObject DetectionSelf-Supervised Learning+1

ParaDiS: Parallelly Distributable Slimmable Neural Networks

2021-10-06 · Alexey Ozerov, Anne Lambert, Suresh Kirthi Kumaraswamy

When several limited power devices are available, one of the most efficient ways to make profit of these resources, while reducing the processing latency and communication load, is to run in parallel several neural sub-n…

Image Super-ResolutionSuper-Resolution

MoSE: Mixture of Slimmable Experts for Efficient and Adaptive Language Models

2026-02-05 · Nurbek Tastan, Stefanos Laskaridis, Karthik Nandakumar, Samuel Horvath arxiv

Mixture-of-Experts (MoE) models scale large language models efficiently by sparsely activating experts, but once an expert is selected, it is executed fully. Hence, the trade-off between accuracy and computation in an Mo…

RTF-Q: Efficient Unsupervised Domain Adaptation with Retraining-free Quantization

2024-08-11 · Nanyang Du, Chen Tang, Yuxiao Jiang, Yuan Meng 외

Performing unsupervised domain adaptation on resource-constrained edge devices is challenging. Existing research typically adopts architecture optimization (e.g., designing slimmable networks) but requires expensive trai…

Domain AdaptationQuantizationUnsupervised Domain Adaptation

Universally Slimmable Networks and Improved Training Techniques

2019-03-12 · ICCV 2019 10 · Jiahui Yu, Thomas Huang

Slimmable networks are a family of neural networks that can instantly adjust the runtime width. The width can be chosen from a predefined widths set to adaptively optimize accuracy-efficiency trade-offs at runtime. In th…

Deep Reinforcement LearningImage Super-ResolutionReinforcement LearningSuper-Resolution