日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

週刊ニュースレター購読
arXiv:1705.05116

Tuning Modular Networks with Weighted Losses for Hand-Eye Coordination

Tuning Modular Networks with Weighted Losses for Hand-Eye Coordination

シェア:XThreadsFacebookLINEはてブBluesky

著者: Fangyi Zhang, Jürgen Leitner, Michael Milford, Peter I. Corke

分類: cs.RO, cs.AI, cs.CV, cs.LG, cs.SY, eess.SY

原文アブストラクト

This paper introduces an end-to-end fine-tuning method to improve hand-eye coordination in modular deep visuo-motor policies (modular networks) where each module is trained independently. Benefiting from weighted losses, the fine-tuning method significantly improves the performance of the policies for a robotic planar reaching task.