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 2210.01007cs.LGcs.AI
收藏

用树进行奖励学习:方法与评价

Reward Learning with Trees: Methods and Evaluation

University of Bristol(布里斯托大学);Thales(泰雷兹集团)

Tom Bewley, Jonathan Lawry, Arthur Richards, Rachel Craddock, Ian Henderson

摘要:

最近从人类反馈中学习奖励函数的努力倾向于使用深度神经网络,其缺乏透明度阻碍了我们解释代理行为或验证对齐的能力。我们探索学习本质上可解释的树模型的优点。我们开发了一个最近提出的方法学习奖励树的偏好标签,并表明它是广泛的竞争对手神经网络具有挑战性的高维任务,具有良好的鲁棒性有限或损坏的数据。在发现奖励树学习可以在复杂的环境中有效地完成之后,我们考虑了为什么应该使用它,证明了可解释的奖励结构为可追溯性,验证和解释提供了重要的范围。

英文摘要:

Recent efforts to learn reward functions from human feedback have tended to use deep neural networks, whose lack of transparency hampers our ability to explain agent behaviour or verify alignment. We explore the merits of learning intrinsically interpretable tree models instead. We develop a recently proposed method for learning reward trees from preference labels, and show it to be broadly competitive with neural networks on challenging high-dimensional tasks, with good robustness to limited or corrupted data. Having found that reward tree learning can be done effectively in complex settings, we then consider why it should be used, demonstrating that the interpretable reward structure gives significant scope for traceability, verification and explanation.

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。

↑