Tuning Modular Networks with Weighted Losses for Hand-Eye Coordination
arXiv:1705.05116v13 citations
Originality Synthesis-oriented
AI Analysis
This work addresses a specific problem in robotics for enhancing modular network policies, but it appears incremental as it fine-tunes existing methods.
The paper tackled improving hand-eye coordination in modular deep visuo-motor policies by introducing an end-to-end fine-tuning method with weighted losses, resulting in significant performance gains for a robotic planar reaching task.
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.