Teaching

Fall 2025: ECE 598 RE - Dynamical Systems and Neural Networks

This graduate-level course explores the deep interplay between dynamical systems theory and neural networks. We move beyond treating neural networks as “black boxes” to understand why they work, using the tools of chaos, stability, and attractor dynamics. The course emphasizes hands-on computational analysis to build a strong theoretical and intuitive understanding of learning and computation in brains and machines.

• Visit the Course Website for details »

Time & Location: Tue/Thu 11:00–12:20 • ECEB 4070
Office Hours: CSL 314, time TBD

Course Description

In this course, we will analyze how stability, attractors, bifurcations, and chaos govern the behavior and learning capabilities of neural networks. The course covers foundational discrete and continuous-time dynamical systems, applying them to the analysis of modern machine learning models (including training dynamics) and complex neural circuits (both rate-based and spiking). The course culminates in a research project applying these interdisciplinary concepts.

Prerequisites

• Required: Multivariate calculus, linear algebra, and basic proficiency in Python or Julia.
• Recommended: Familiarity with ordinary differential equations and foundational machine learning concepts.