Chenguang Li

This paper introduces a theory where policy optimization emerges from reward-modulated noise in a model of representation learning.

This paper demonstrates that neural networks can exhibit reward-seeking behavior through local predictive updates and internal noise, enabling autonomous and adaptable interaction with environments.

This paper demonstrates a hybrid system combining optogenetics and deep reinforcement learning to enhance nematode navigation and food search performance.

This paper shows that NMDA receptors improve the consistency of synaptic integration by working with AMPA receptors to reduce variability in neuronal responses.