arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2101.04750cs.LGcs.AIstat.ML
收藏

基于线性表示的Meta-强化学习即时适应算法

Linear Representation Meta-Reinforcement Learning for Instant Adaptation

University of California, Berkeley(加州大学伯克利分校)

Matt Peng, Banghua Zhu, Jiantao Jiao

摘要:

本文介绍了快速线性化自适应策略(FLAP),这是一种新的Meta强化学习(元RL)方法,能够很好地外推到分布外的任务,而不需要重用训练数据,并且在测试过程中只需要少量样本就可以几乎立即适应。FLAP建立在学习策略的共享线性表示的思想之上,以便在适应新任务时,预测一组线性权重就足够了。一个单独的适配器网络与策略同时训练,这样在适配过程中,我们可以直接使用适配器网络来预测这些线性权重,而不是通过梯度下降来更新元策略,例如在MAML等现有的元RL方法中,以获得新策略。单独的前馈网络的应用不仅显著加快了自适应运行时间,而且还非常好地推广到以前的Meta-RL方法无法推广的非常不同的任务。在标准连续控制元RL基准测试上的实验表明,与现有方法相比,FLAP在分布外任务上表现出更强的性能,平均回报率高达两倍,自适应运行速度高达8倍。

英文摘要:

This paper introduces Fast Linearized Adaptive Policy (FLAP), a new meta-reinforcement learning (meta-RL) method that is able to extrapolate well to out-of-distribution tasks without the need to reuse data from training, and adapt almost instantaneously with the need of only a few samples during testing. FLAP builds upon the idea of learning a shared linear representation of the policy so that when adapting to a new task, it suffices to predict a set of linear weights. A separate adapter network is trained simultaneously with the policy such that during adaptation, we can directly use the adapter network to predict these linear weights instead of updating a meta-policy via gradient descent, such as in prior meta-RL methods like MAML, to obtain the new policy. The application of the separate feed-forward network not only speeds up the adaptation run-time significantly, but also generalizes extremely well to very different tasks that prior Meta-RL methods fail to generalize to. Experiments on standard continuous-control meta-RL benchmarks show FLAP presenting significantly stronger performance on out-of-distribution tasks with up to double the average return and up to 8X faster adaptation run-time speeds when compared to prior methods.

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。

↑