I study the relationship between machine learning and dynamical systems, with applications to neuroscience and biology more generally. A major current focus is the dynamics of biological networks, particularly oscillator networks. Other recent work has explored machine learning methods for optimally controlling neural dynamics and how these methods can be used to understand the cortical processes underlying higher cognition. Here is an overview of ongoing work:
Learning to control
biological network dynamics
Given a population of networked agents with random features, how do we
connect them to achieve an optimal dynamics?

Related Work:
  • Ricci, M.G., Zhang, Y., Soni, A., Jung, M., & Serre, T. Learning to control huge systems of coupled oscillators (In preparation)
  • Ricci, M.G., Zhang, Y., Soni, A., Jung, M., & Serre, T. Kura-Net: Exploring systems of coupled oscillators using deep learning. COSYNE, 2020.
  • Ricci, M.G., Windolf, C., & M., Serre, T., A Formal Model of Neural Synchrony for Unsupervised Image Grouping. COSYNE, 2019.
Neural optimal control
Can we use principles from optimal control theory to learn adaptive policies
for steering dynamical systems?

Related Work:
Understanding the neural dynamics
of visual reasoning
How do machines reason about images compared to humans, and what
principles of neural dynamics underlie this ability?

Related Work:
Network analysis of
animal disease models
What can graph theory and the theory of complex networks tell us about
animal behavior and disease? This is joint work with the lab of Victoria Abraira .

Related Work:
  • Ricci, M.G., Thackray, J., Theis, T., Abraira, V. Automated modeling of higher-order behavioral sequences in mouse models of spinal injury (In preparation)