FOCUS: 最適輸送による教師なし模倣のための一階制御
FOCUS: First-Order Control Using Optimal Transport Supervision for Imitation from Observations
観測のみからの模倣学習において、局所的な最適輸送マッチングと短ホライズンの微分可能アクター・クリティック学習を組み合わせ、接触の多い筋骨格ヒューマノイド制御でも安定かつ高効率に学習できる手法を提案した。
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: João A. Cândido Ramos, Lionel Blondé, Hugues Vinzant, Jacopo Castellini, Alexandros Kalousis
分類: cs.LG, cs.RO
原文アブストラクト
Imitation learning from observations must recover behavior from state-only demonstrations without known temporal correspondence. Differentiable simulators provide pathwise policy gradients, but differentiating through long, contact-rich trajectories can make optimization unstable. We introduce FOCUS, which combines local optimal-transport alignment with short-horizon differentiable actor-critic learning. FOCUS matches agent rollout windows to multiple sampled expert segments using a soft best-of-$K$ objective that emphasizes locally compatible matches. The resulting transport-based costs provide differentiable imitation rewards, while a learned critic bootstraps value beyond the rollout horizon. FOCUS requires neither temporally aligned trajectories nor expert-aligned initialization or teacher forcing. Across contact-rich locomotion tasks in DFlex, including a musculoskeletal humanoid controlled by 152 muscle activations, FOCUS improves sample efficiency and training stability over adversarial imitation-from-observation methods and prior differentiable imitation baselines.