| 8:30 AM – 9:00 AM | Pre-workshop networking
|
| 9:00 AM - 9:10 AM | Opening Remarks
|
| 9:15 AM - 10:00 AM | Invited Talk 1: Neural Certificates in Large-Scale Autonomy Design (Chuchu Fan)
Abstract: Learning-enabled control systems have demonstrated impressive empirical performance on challenging control problems in robotics, but this performance comes at the cost of reduced transparency and lack of guarantees on the safety or stability of the learned controllers. In recent years, new techniques have emerged to provide these guarantees by learning certificates alongside control policies. These certificates provide concise, data-driven proofs that guarantee the safety and stability of the learned control system. These methods allow the user to verify the safety of a learned controller and provide supervision during training, allowing safety and stability requirements to influence the training process. This talk presents an overview of this rapidly developing field of certificate learning. We hope that this talk will serve as an accessible introduction to the theory and practice of certificate learning, both to those who wish to apply these tools to practical robotics problems and to those who wish to dive more deeply into the theory of learning for control.
|
| 10:10 AM – 10:20 AM | Contributed Talk 1: A Coupled Variational Encoder-Decoder - DeepONet Surrogate Model for the Rayleigh-Bénard Convection Problem
|
| 10:25 AM – 10:35 AM | Contributed Talk 2: Why does SGD Prefer Flat Minima?: Through the Lens of Dynamical Systems
|
| 10:40 AM – 10:50 AM | Contributed Talk 3: Persistence-Based Discretization for Learning Discrete Event Systems from Time Series
|
| 10:55 AM – 11:05 AM | Contributed Talk 4: Multimodal Teacher Forcing for Reconstructing Nonlinear Dynamical Systems
|
| 11:15 AM - 12:00 PM | Invited Talk 2: The Future of Governing Equations (J. Nathan Kutz)
Abstract: Machine learning and AI algorithms are transforming a diverse number of fields in science and engineering. This is largely due their success in model discovery which turns data into reduced order models and neural network representations that are not just predictive, but provide insight into the nature of the underlying dynamical system that generated the data. We introduce a number of data-driven strategies for discovering nonlinear multiscale dynamical systems, compact representations, and their embeddings from data. Importantly, data-driven architectures must jointly discover coordinates and parsimonious models in order to produce maximally generalizable and interpretable models of physics-based systems and processes.
|
| 12:00 PM - 2:00 PM | Lunch Break
|
| 2:10 PM – 2:55 PM | Invited Talk 3: From Online Convex Optimization to Control and Back (Elad Hazan)
Abstract: In this talk we will discuss an emerging paradigm in differentiable reinforcement learning called “online nonstochastic control”. The new approach applies techniques from online convex optimization and convex relaxations to obtain new methods with provable guarantees for classical settings in optimal and robust control. Time permitting we will discuss recent extensions to nonlinear adaptive control and planning, and a recent application to meta-optimization. No background is required for this talk, and relevant materials can be found here.
|
| 3:05 PM – 3:15 PM | Contributed Talk 5: Stochastic Bilevel Projection-Free Optimization
|
| 3:20 PM – 3:30 PM | Contributed Talk 6: Dynamic Function Learning through Control of Ensemble Systems
|
| 3:35 PM – 3:45 PM | Contributed Talk 7: Occupancy Information Ratio: Infinite-Horizon, Information-Directed, Parameterized Policy Search
|
| 3:50 PM – 4:00 PM | Contributed Talk 8: SpReME: Sparse Regression for Multi-Environment Dynamic Systems
|
| 4:00 PM – 5:00 PM | Poster Session
|
| 5:00 PM - 5:10 PM | Closing Remarks
|
| 5:10 PM - 6:00 PM | Post-workshop networking
|