ハイブリッド制御のための連続・離散強化学習
Continuous-Discrete Reinforcement Learning for Hybrid Control in Robotics
離散行動と連続行動が混在するロボット制御問題を、両者を同時に最適化するハイブリッド強化学習でネイティブに解き、専門家ヒューリスティクスを排除する利点をシミュレーションと実機で示した論文。
著者: Michael Neunert, Abbas Abdolmaleki, Markus Wulfmeier, Thomas Lampe, Jost Tobias Springenberg, Roland Hafner, Francesco Romano, Jonas Buchli, Nicolas Heess, Martin Riedmiller
分類: cs.LG, cs.RO, stat.ML
原文アブストラクト
Many real-world control problems involve both discrete decision variables - such as the choice of control modes, gear switching or digital outputs - as well as continuous decision variables - such as velocity setpoints, control gains or analogue outputs. However, when defining the corresponding optimal control or reinforcement learning problem, it is commonly approximated with fully continuous or fully discrete action spaces. These simplifications aim at tailoring the problem to a particular algorithm or solver which may only support one type of action space. Alternatively, expert heuristics are used to remove discrete actions from an otherwise continuous space. In contrast, we propose to treat hybrid problems in their 'native' form by solving them with hybrid reinforcement learning, which optimizes for discrete and continuous actions simultaneously. In our experiments, we first demonstrate that the proposed approach efficiently solves such natively hybrid reinforcement learning problems. We then show, both in simulation and on robotic hardware, the benefits of removing possibly imperfect expert-designed heuristics. Lastly, hybrid reinforcement learning encourages us to rethink problem definitions. We propose reformulating control problems, e.g. by adding meta actions, to improve exploration or reduce mechanical wear and tear.
