強化学習/メタ学習arXiv:2007.08616
メタ学習による衝突回避ロボティクス
Collision Avoidance Robotics Via Meta-Learning (CARML)
LIDAR搭載2D車両の障害物回避と目標到達を多目的強化学習問題として定式化し、MAMLで学習する手法を提案、TD3ベースラインと比較した。
著者: Abhiram Iyer, Aravind Mahadevan
分類: cs.LG, cs.AI, cs.RO, stat.ML
原文アブストラクト
This paper presents an approach to exploring a multi-objective reinforcement learning problem with Model-Agnostic Meta-Learning. The environment we used consists of a 2D vehicle equipped with a LIDAR sensor. The goal of the environment is to reach some pre-determined target location but also effectively avoid any obstacles it may find along its path. We also compare this approach against a baseline TD3 solution that attempts to solve the same problem.
関連論文
- 効率的探索のための教師なし学習:自己設定目標による適応的方策の事前学習強化学習/メタ学習
