Explore cutting-edge research at the intersection of machine learning and dynamical systems through this comprehensive symposium featuring expert presentations on climate prediction, reservoir computing, chaos control, and data-driven modeling. Delve into advanced topics including Koopman operator theory, neural differential equations, transfer operator learning, and generative modeling approaches for complex dynamical systems. Discover innovative applications spanning weather forecasting, cancer progression modeling, and robust network reconstruction while examining theoretical foundations of operator learning, statistical estimation, and optimization in dynamical contexts. Learn about state-of-the-art methodologies for system identification, control theory, and prediction in chaotic systems, with particular emphasis on data-driven approaches that bridge traditional mathematical modeling with modern machine learning techniques. Gain insights into emerging areas such as physics-informed neural networks, score-based diffusion models, and equivariant learning theory as applied to dynamical systems analysis and prediction.

Syllabus

Machine Learning for Prediction of Earth Climate and Weather
Understanding forecasting with reservoir computing via synchronization
Kernelization of Reservoir Systems
Controlling Chaos Using Edge Computing Hardware
Predicting tipping point with reservoir computing
Training Autonomous Dynamics of a Soft Body: Embedding Bifurcation Structures...
State estimation of complex systems
Learning, approximation and control
Linear Operator Theoretic Framework for Data-Driven Optimal Control:
Several topics at the intersection of control, dynamics, and learning from data
The Operator is the Model
Koopman-based generalization bound for neural networks
On the Barriers of Robust Koopman Learning
Statistical Learning of Transfer Operators and the Infinitesimal Generator
Representation Learning for Dynamical Systems
Combinatorial Topological Dynamics
Identifying nonlinear dynamics with high confidence from sparse data
On finite-dimensional approximations of push-forwards on locally analytic functionals
Closed-Loop Koopman Operator Approximation
Low-dimensional approximations of the conditional law of Volterra processes...
Data-driven reduced order models of forced systems using invariant foliations
Should we Derive or Let the Data Drive? Symbiotizing Data-driven Learning...
Some older, and some current, thoughts on Data and the Modeling of Complex Systems
Operator Learning Without the Adjoint
Learning Port Hamiltonian structures using PINNs type architecture
Provable Posterior Sampling with Score-Based Diffusion through Tilted Transport
Learning and Dynamical Systems: Perspectives from Optimization, Control, and Robotics
Learning Coarse-Grained Dynamics on Graph
Differentiable Programming for Data-driven Modeling, Optimization, and Control
Detecting non-trivial cycles of point clouds and time series data on manifolds
Exploring Cancer Progression: From Static Imaging Data to System Dynamics
Non-smooth dynamics and machine learning
Avoidance of traps for nonconvex stochastic optimization and equilibrium learning in games
Non-Euclidean Generative Modeling
A dynamical systems perspective on measure transport and generative modeling
Phase Transition Theory fo the Score Degradation of Machine Learning Models
Dynamical systems in deep generative modelling
Discovering dynamics and parameters of nonlinear systems from partial observations
Gated Recurrent Neural Networks with Weighted Time-Delay Feedback
Continuum Attention for Neural Operators
Ergodic Basis Pursuit induces robust sparse network reconstruction
Anamnesic Neural Differential Equations with Orthogonal Polynomial Projections
Learning Transfer Operators by Kernel Density Estimation
Graphs of Random Matrices in Deep Learning
Data-adaptive RKHS regularization for learning kernels in operators
Ensemble forecasts in reproducing kernel Hilbert space family
Equivariant learning through invariant theory
Ensemble forecasts in reproducing kernel Hilbert space family
Orbit hierarchies and long-term predictability in chaotic systems
Simplicity bias, algorithmic probability, and time series
On Bridging Machine Learning, Dynamical Systems, and Algorithmic Information Theory