新しい行動集合へのゼロショット汎化を実現する強化学習フレームワーク
Generalization to New Actions in Reinforcement Learning
行動情報から行動表現を推論し、未知の行動集合にも柔軟に対応できる方策を二段階で学習する手法を提案し、物理パズルや積み木タスクで評価した。
著者: Ayush Jain, Andrew Szot, Joseph J. Lim
分類: cs.LG, cs.AI, cs.RO, stat.ML
原文アブストラクト
A fundamental trait of intelligence is the ability to achieve goals in the face of novel circumstances, such as making decisions from new action choices. However, standard reinforcement learning assumes a fixed set of actions and requires expensive retraining when given a new action set. To make learning agents more adaptable, we introduce the problem of zero-shot generalization to new actions. We propose a two-stage framework where the agent first infers action representations from action information acquired separately from the task. A policy flexible to varying action sets is then trained with generalization objectives. We benchmark generalization on sequential tasks, such as selecting from an unseen tool-set to solve physical reasoning puzzles and stacking towers with novel 3D shapes. Videos and code are available at https://sites.google.com/view/action-generalization
