arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1802.08802cs.AI
收藏

Reinforcement Learning on Web Interfaces Using Workflow-Guided Exploration

Stanford University（斯坦福大学）

Evan Zheran Liu, Kelvin Guu, Panupong Pasupat, Tianlin Shi, Percy Liang

备注：
International Conference on Learning Representations (ICLR), 2018
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

英文摘要：

Reinforcement learning (RL) agents improve through trial-and-error, but when reward is sparse and the agent cannot discover successful action sequences, learning stagnates. This has been a notable problem in training deep RL agents to perform web-based tasks, such as booking flights or replying to emails, where a single mistake can ruin the entire sequence of actions. A common remedy is to "warm-start" the agent by pre-training it to mimic expert demonstrations, but this is prone to overfitting. Instead, we propose to constrain exploration using demonstrations. From each demonstration, we induce high-level "workflows" which constrain the allowable actions at each time step to be similar to those in the demonstration (e.g., "Step 1: click on a textbox; Step 2: enter some text"). Our exploration policy then learns to identify successful workflows and samples actions that satisfy these workflows. Workflows prune out bad exploration directions and accelerate the agent's ability to discover rewards. We use our approach to train a novel neural policy designed to handle the semi-structured nature of websites, and evaluate on a suite of web tasks, including the recent World of Bits benchmark. We achieve new state-of-the-art results, and show that workflow-guided exploration improves sample efficiency over behavioral cloning by more than 100x.

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。

↑