This repository contains code that implements the ODIN algorithm as presented in the paper:
ODIN: ODE-Informed Regression for Parameter and State Inference in Time-Continuous Dynamical Systems. Philippe Wenk*, Gabriele Abbati*, Michael A Osborne, Bernhard Schölkopf, Andreas Krause and Stefan Bauer. Proceedings of the 34th AAAI Conference on Artificial Intelligence (AAAI 20). New York, NY, USA. AAAI Press.
*: equal contribution
The code provided is written in Python 3.6, and relies on the following libraries:
Some usage examples (the same ones used in the experimental section of the paper) can be found in the examples/ directory. Comments in the code illustrate how it can be further applied to generic ODE models not included here.