Ben Day
Papers
1
Total Citations
71
H-Index
1
About
Ben Day is a researcher at the intersection of reinforcement learning (RL) and evolutionary computation, with a focus on scaling biologically inspired algorithms to deep neural networks. His most cited work, "Proximal Distilled Evolutionary Reinforcement Learning" (2020, 71 citations), introduces a novel hybrid framework that combines the exploration strengths of genetic algorithms with the sample efficiency of RL. By distilling evolutionary search into a policy network, Day’s approach bridges a long-standing gap between these two paradigms, enabling GAs to tackle complex, high-dimensional environments previously dominated by RL. This work has been influential in advancing multi-agent systems and robotics, where robust exploration is critical. Day’s contributions highlight how evolutionary methods can be modernized for deep learning, offering a complementary path to traditional RL. With growing recognition in the field, his research continues to shape discussions on algorithm hybridization, making him a key voice for students and researchers interested in the future of adaptive, scalable learning systems.
Research Focus
Key Achievements
Top Papers
- 1Proximal Distilled Evolutionary Reinforcement Learning71 citations · 2020