Dual-Force: 模倣制約下でのオフライン多様性最大化の強化
Dual-Force: Enhanced Offline Diversity Maximization under Imitation Constraints
模倣制約下でオフラインデータから多様な行動を学習する手法を提案し、ファンデルワールス力を用いた多様性推定と事前学習済み報酬エンコーディングで安定化を図り、Solo12の歩行と障害物回避で有効性を示した。
著者: Pavel Kolev, Marin Vlastelica, Georg Martius
分類: cs.LG, cs.AI, cs.RO
原文アブストラクト
Offline diversity maximization under imitation constraints can transform demonstration data into a set of distinct behavioral policies, improving robustness to distribution shift without additional environment interaction. In practice, however, existing offline approaches often rely on mutual-information objectives that require training a skill discriminator and can become unstable under the non-stationary rewards induced by alternating Lagrangian optimization. We introduce Dual-Force, an offline algorithm that (i) maximizes diversity using an off-policy estimator of a Van der Waals (VdW) force objective computed from successor features, eliminating the skill discriminator, and (ii) stabilizes training under non-stationary intrinsic rewards by conditioning the value function and policy on a pre-trained Functional Reward Encoding (FRE). The FRE code also enables zero-shot recall of every encountered skill via its associated latent representation, removing the need to pre-specify a fixed number of skills. On two Solo12 simulation benchmarks (locomotion and obstacle navigation), Dual-Force recovers diverse high-performing behaviors while matching a target expert state occupancy and improves robustness in adversarial obstacle variations.
