双対優位フィールド:オフライン目標条件付き強化学習のための方策抽出法
Dual Advantage Fields
双対線形価値モデルから局所的な優位信号を抽出する方策抽出法を提案し、オフライン目標条件付き強化学習の性能を向上させた。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Alexey Zemtsov, Maxim Bobrin, Alexander Nikulin, Dmitry V. Dylov, Fakhri Karray, Vladislav Kurenkov, Martin Takáč, Arip Asadulaev
分類: cs.LG, cs.AI, cs.RO
原文アブストラクト
Offline goal-conditioned reinforcement learning requires both long-horizon reachability estimates and local action comparisons. Dual goal representations provide value fields that capture global goal reachability, but they do not directly specify which action should be preferred at a given state. We propose Dual Advantage Fields, a policy-extraction method that turns a bilinear dual value model into a local advantage signal. Under bilinear dual parameterization, the goal embedding is the gradient of the value field with respect to the state representation. DAF learns an action-effect model that predicts the discounted feature displacement induced by an action and scores actions by the alignment between this displacement and the goal direction. In the realizable case, this score equals the goal-conditioned Bellman advantage, yielding a standard local policy-improvement guarantee. On OGBench locomotion, manipulation, and puzzle tasks, DAF improves aggregate RLiable metrics and performs strongly in settings where locally correct actions differ from direct movement toward the final goal.
