Contraction Theory for Machine Learning
A Tutorial Overview
How can we mathematically ensure the safety, stability, and robustness of machine learning-based control and estimation systems?
This website provides a tutorial overview of contraction theory for nonlinear stability analysis and control synthesis of deterministic and stochastic systems, with an emphasis on deriving formal robustness and stability guarantees for various learning-based control problems.
This site is maintained by Soon-Jo Chung (sjchung@caltech.edu) and Jean-Jacques Slotine (jjs@mit.edu) with help from Hiroyasu Tsukamoto (htsukamoto@caltech.edu).
Contraction Theory Tutorial Papers for Beginners
The original paper that derives contraction theory for nonlinear incremental stability analysis
A tutorial paper on utilizing contraction theory for learning-based control with formal robustness and stability guarantees
Tutorial Session at 60th IEEE Conference on Decision and Control (CDC)
Safe Motion Planning with Tubes and Contraction Metrics
S. Singh (Google), H. Tsukamoto (Caltech), B. Lopez (UCLA), S.-J. Chung (Caltech), J.-J. Slotine (MIT)
Coming Soon
I. R. Manchester (U of Sydney), M. Revay (U of Sydney), R. Wang (U of Sydney)