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 1902.09434cs.LGstat.ML
收藏

S-TRIGGER:基于自触发生成重放的连续状态表示学习

S-TRIGGER: Continual State Representation Learning via Self-Triggered Generative Replay

ENSTA Paris(巴黎高等先进技术学院);Institut Polytechnique de Paris(巴黎理工学院);INRIA(法国国家信息与自动化研究所);City University of London(伦敦城市大学)

Hugo Caselles-Dupré, Michael Garcia-Ortiz, David Filliat

摘要:

我们考虑在持续学习环境中建立控制的状态表示模型的问题。随着环境的变化,目标是在不丢失过去知识的情况下有效地压缩感知状态的信息,然后对所得特征使用强化学习进行有效的策略学习。为此,我们提出了S-TRIGGER,一种适用于变分自动编码器及其许多变体的连续状态表示学习的通用方法。该方法基于生成重放,即使用生成的样本来维护过去的知识。它附带了一个统计上合理的环境变化检测方法,可以自动触发生成重放。我们在VAE上的实验表明,S-TRIGGER学习状态表示,允许快速和高性能的强化学习,同时避免灾难性遗忘。由此产生的系统能够自主学习新的信息,而不使用过去的数据和有限的系统规模。我们的实验代码附在附录中。

英文摘要:

We consider the problem of building a state representation model for control, in a continual learning setting. As the environment changes, the aim is to efficiently compress the sensory state's information without losing past knowledge, and then use Reinforcement Learning on the resulting features for efficient policy learning. To this end, we propose S-TRIGGER, a general method for Continual State Representation Learning applicable to Variational Auto-Encoders and its many variants. The method is based on Generative Replay, i.e. the use of generated samples to maintain past knowledge. It comes along with a statistically sound method for environment change detection, which self-triggers the Generative Replay. Our experiments on VAEs show that S-TRIGGER learns state representations that allows fast and high-performing Reinforcement Learning, while avoiding catastrophic forgetting. The resulting system is capable of autonomously learning new information without using past data and with a bounded system size. Code for our experiments is attached in Appendix.

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。

↑