Andrew Mundy
Papers
1
Total Citations
36
H-Index
1
About
Andrew Mundy’s research sits at the intersection of neuromorphic engineering, robotics, and unsupervised learning, where he explores how biological principles can inspire more adaptive, energy-efficient machines. His most cited work, “Serendipitous Offline Learning in a Neuromorphic Robot” (2016, 36 citations), introduces a hybrid learning paradigm that enables a mobile robot to acquire complex sensorimotor mappings from a small set of hard-coded reflex behaviors. By feeding all sensor data through a spike-based silicon retina, Mundy demonstrates how offline, serendipitous learning—where the robot’s own reflexive actions generate useful training signals—can bootstrap sophisticated behaviors without explicit supervision. This contribution is notable for bridging the gap between hand-designed control and autonomous adaptation in neuromorphic systems. Mundy’s work has influenced researchers seeking to reduce the energy and data requirements of robotic learning, and his approach remains a touchstone for those building spike-based, self-organizing robots. His achievements highlight a deep commitment to merging computational neuroscience with practical robotics, offering a path toward machines that learn as efficiently as biological organisms.
Research Focus
Key Achievements
Top Papers
- 1Serendipitous Offline Learning in a Neuromorphic Robot36 citations · 2016