Dynamical Intelligence Group (DIG)
Research Group @ Johns Hopkins University Electrical and Computer Engineering Department
DIG is an interdisciplinary group of researchers and students who are interested in understanding the dynamics of learning and the learning of dynamics in artificial and biological systems. We work in the intersection of dynamical systems and control theory, machine learning, and neuroscience.
news
| Apr 23, 2026 | DIG is excited to have been awarded an International Brain Research Organization (IBRO) Collaborative Research Grant funded to expand our collaboration with Prof. Roberto Bottini’s group (University of Trento)! We appreciate the support. |
|---|---|
| Mar 28, 2026 | New preprint out on using Koopman operator theory for characterizing reinforcement learning behavior: Interpreting Reinforcement Learning Model Behavior via Koopman with Control ! |
| Mar 12, 2026 | DIG is very excited to have Edouard de Ponnat joining us in the Fall as an ECE graduate student - welcome Edouard! |
| Dec 19, 2025 | New preprint out on grid cell computation: Distance by de-correlation: Computing distance with heterogeneous grid cells. A great collaboration with Priti Dasbehera and Akshunna S. Dogra! |
selected publications
- ChaosKoopman learning with episodic memoryChaos: An Interdisciplinary Journal of Nonlinear Science, 2025
- NeurIPSIdentifying equivalent training dynamicsAdvances in Neural Information Processing Systems, 2024
- NeurIPSNot so griddy: Internal representations of RNNs path integrating more than one agentAdvances in Neural Information Processing Systems, 2024