Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster B in Poster Session B: Tuesday, August 4, 2:00 – 3:45 pm, Kimmel Center, Shorin & Rosenthal Rooms

Heterogeneous Time Constants Enhance Robustness in Equilibrium Propagation

Yoshimasa Kubo¹, Suhani Pragnesh Modi¹, Smit Patel¹; ¹Lakehead University

Presenter: Yoshimasa Kubo

Equilibrium propagation (EP) is a biologically plausible alternative to backpropagation for training neural networks. However, existing EP models typically use a uniform scalar integration step dt, whereas biological neurons exhibit heterogeneous temporal dynamics associated with diverse membrane time constants. Here, we introduce heterogeneous time steps (HTS) for EP by assigning neuron-specific temporal scales drawn from biologically motivated distributions. We show that HTS maintains competitive task performance while providing modest improvements in learning robustness. These results suggest that incorporating heterogeneous temporal dynamics enhances both the biological realism and robustness of equilibrium propagation.

Topic Area: Methods, Tools, Theory & Neural Coding