動力学ランダム化はいつ文脈推論を試すのか
When Does Dynamics Randomization Test Context Inference?
ランダム化された動力学環境で文脈推論の効果を測る際、状態と行動から得られる予測情報の余裕(predictive headroom)を定式化し、その大きさが実験結果の解釈に与える影響を分析した論文。
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: Oleg Smirnov, Sofiane Ennadir, John Pertoft, Bjartur Hjaltason, Sara Karimi
分類: cs.LG
原文アブストラクト
Environments with randomized dynamics are widely used to ask whether world models need explicit context inference. However, randomization can leave little predictive information for a context encoder to add, because distinct latent parameters may produce similar transitions at visited states, or those states may already reveal the active context. We formalize the remaining one-step information as predictive headroom and compare kernels under fixed and visited state--action distributions to distinguish changes in query coverage from changes in context weights. We analyze discrete and continuous environments and find low headroom both when kernels differ little and when visitation reduces an otherwise large discrepancy. At the same time, resetting state components increases headroom, showing that context evidence accumulates in state. We propose reporting this headroom alongside the randomized parameters, so that gains from explicit context inference can be interpreted relative to the information missing from the current state and action.