arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1805.06370stat.MLcs.LG
收藏

Progress & Compress:一个可扩展的持续学习框架

Progress & Compress: A scalable framework for continual learning

DeepMind(深度思维);University of Oxford(牛津大学)

Jonathan Schwarz, Jelena Luketina, Wojciech M. Czarnecki, Agnieszka Grabska-Barwinska, Yee Whye Teh, Razvan Pascanu, Raia Hadsell

摘要:

我们介绍了一个概念上简单和可扩展的框架,连续学习领域的任务是按顺序学习。我们的方法在参数数量上是恒定的,旨在保持先前遇到的任务的性能,同时加快后续问题的学习进度。这是通过训练一个具有两个组件的网络来实现的:一个知识库,能够解决以前遇到的问题,它连接到一个活动列,用于有效地学习当前任务。在学习了一项新任务后,活动列被提取到知识库中,注意保护任何以前获得的技能。这种主动学习(进展)和整合(压缩)的循环不需要架构增长,不需要访问或存储以前的数据或任务,也不需要特定于任务的参数。我们展示了手写字母顺序分类的进步和压缩方法以及两个强化学习领域:Atari游戏和3D迷宫导航。

英文摘要:

We introduce a conceptually simple and scalable framework for continual learning domains where tasks are learned sequentially. Our method is constant in the number of parameters and is designed to preserve performance on previously encountered tasks while accelerating learning progress on subsequent problems. This is achieved by training a network with two components: A knowledge base, capable of solving previously encountered problems, which is connected to an active column that is employed to efficiently learn the current task. After learning a new task, the active column is distilled into the knowledge base, taking care to protect any previously acquired skills. This cycle of active learning (progression) followed by consolidation (compression) requires no architecture growth, no access to or storing of previous data or tasks, and no task-specific parameters. We demonstrate the progress & compress approach on sequential classification of handwritten alphabets as well as two reinforcement learning domains: Atari games and 3D maze navigation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑