エージェント設計のための拡張可能な対話型インターフェース
An Extensible Interactive Interface for Agent Design
複雑なタスクを報酬関数で指定する代わりに、ユーザーが既存ポリシーを切り替えながらデモンストレーションを行い、それに基づいて新ポリシーを学習する対話型インターフェースを提案。Lunar Landerで有効性を示した。
著者: Matthew Rahtz, James Fang, Anca D. Dragan, Dylan Hadfield-Menell
分類: cs.LG, cs.HC, cs.RO, stat.ML
原文アブストラクト
In artificial intelligence, we often specify tasks through a reward function. While this works well in some settings, many tasks are hard to specify this way. In deep reinforcement learning, for example, directly specifying a reward as a function of a high-dimensional observation is challenging. Instead, we present an interface for specifying tasks interactively using demonstrations. Our approach defines a set of increasingly complex policies. The interface allows the user to switch between these policies at fixed intervals to generate demonstrations of novel, more complex, tasks. We train new policies based on these demonstrations and repeat the process. We present a case study of our approach in the Lunar Lander domain, and show that this simple approach can quickly learn a successful landing policy and outperforms an existing comparison-based deep RL method.
