期刊
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
卷 -, 期 -, 页码 4441-4451出版社
IEEE COMPUTER SOC
DOI: 10.1109/CVPR46437.2021.00442
关键词
-
资金
- MIAI@Grenoble Alpes [ANR-19-P3IA-0003]
Standard learning approaches often result in fragile models that are prone to catastrophic forgetting, but we propose a domain randomization technique to build more robust models. By randomizing the distribution of the current domain, our approach shows reduced vulnerability to catastrophic forgetting when transferred to new domains through meta-learning.
Most standard learning approaches lead to fragile models which are prone to drift when sequentially trained on samples of a different nature-the well-known catastrophic forgetting issue. In particular, when a model consecutively learns from different visual domains, it tends to forget the past domains in favor of the most recent ones. In this context, we show that one way to learn models that are inherently more robust against forgetting is domain randomization-for vision tasks, randomizing the current domain's distribution with heavy image manipulations. Building on this result, we devise a meta-learning strategy where a regularizer explicitly penalizes any loss associated with transferring the model from the current domain to different auxiliary meta-domains, while also easing adaptation to them. Such meta-domains are also generated through randomized image manipulations. We empirically demonstrate in a variety of experiments-spanning from classification to semantic segmentation-that our approach results in models that are less prone to catastrophic forgetting when transferred to new domains.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据