2021/08/18 by Joshua Powers, Powers, Joshua, Ryan Grindle +6 · 1 citation
Computer Science · #FOS: Computer and information sciences #Multi-Agent Systems and Negotiation #Neural and Evolutionary Computing (cs.NE) #Optimization and Search Problems #Reinforcement Learning in Robotics #Robotics (cs.RO) #cs.NE #cs.RO
paper · pdf · doi:10.48550/arxiv.2108.08398
arXiv admin note: text overlap with arXiv:1910.07487
openalex publication_date 2021/08/18 · arxiv created 2021/08/20 · arxiv updated 2021/08/24 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In robotics, catastrophic interference continues to restrain policy training across environments. Efforts to combat catastrophic interference to date focus on novel neural architectures or training methods, with a recent emphasis on policies with good initial settings that facilitate training in new environments. However, none of these methods to date have taken into account how the physical architecture of the robot can obstruct or facilitate catastrophic interference, just as the choice of neural architecture can. In previous work we have shown how aspects of a robot's physical structure (specifically, sensor placement) can facilitate policy learning by increasing the fraction of optimal policies for a given physical structure. Here we show for the first time that this proxy measure of catastrophic interference correlates with sample efficiency across several search methods, proving that favorable loss landscapes can be induced by the correct choice of physical structure. We show that such structures can be found via co-optimization -- optimization of a robot's structure and control policy simultaneously -- yielding catastrophic interference resistant robot structures and policies, and that this is more efficient than control policy optimization alone. Finally, we show that such structures exhibit sensor homeostasis across environments and introduce this as the mechanism by which certain robots overcome catastrophic interference.