日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

強化学習arXiv:2404.02728

ロボティクスにおける効果的な行動の教師なし学習

Unsupervised Learning of Effective Actions in Robotics

シェア:XThreadsFacebookLINEはてブBluesky

連続的な動作空間を教師なしで離散化し、環境に効果を生む「行動プロトタイプ」を自動生成する手法を提案。階段昇降の強化学習タスクで、一様・ランダム離散化より速く収束し高い報酬を得た。

著者: Marko Zaric, Jakob Hollenstein, Justus Piater, Erwan Renaudo

分類: cs.RO, cs.AI, cs.LG

原文アブストラクト

Learning actions that are relevant to decision-making and can be executed effectively is a key problem in autonomous robotics. Current state-of-the-art action representations in robotics lack proper effect-driven learning of the robot's actions. Although successful in solving manipulation tasks, deep learning methods also lack this ability, in addition to their high cost in terms of memory or training data. In this paper, we propose an unsupervised algorithm to discretize a continuous motion space and generate "action prototypes", each producing different effects in the environment. After an exploration phase, the algorithm automatically builds a representation of the effects and groups motions into action prototypes, where motions more likely to produce an effect are represented more than those that lead to negligible changes. We evaluate our method on a simulated stair-climbing reinforcement learning task, and the preliminary results show that our effect driven discretization outperforms uniformly and randomly sampled discretizations in convergence speed and maximum reward.

関連論文

  1. 1Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  2. 2GXOとApptronik、Apolloの物流現場試験を拡大 最大1,000台の展開を構想量産・展開
  3. 3newmoとKDDIが資本業務提携 2028年に堺市で自動運転タクシー商用化目指す提携・買収
  4. 4NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  5. 5UC BerkeleyとNVIDIAの研究者、AIエージェントでロボット制御コードを仮想環境で改善研究・技術

PR・アフィリエイト広告

FRONTIER(GPU搭載BTOパソコン)

FRONTIER(GPU搭載BTOパソコン)の公式広告バナーFRONTIERでGPU搭載PCを見る ↗
PR本紙発行元 EmplifAI