OpenDP a free Reinforcement Learning toolbox for discrete time control problems
Résumé
OpenDP (http://opendp.sourceforge.net) is a free software (under GPL) written in C++ for control problems: (i) with discrete time steps (either finite horizon, or approximation of infinite horizon by temporal-ring for stationnary problems), (ii) where the transition function newState = T (state, decision) can be encoded in C++, (iii) possibly stochastic (with the function T above depending on some random process).
Domaines
Apprentissage [cs.LG]Origine | Fichiers produits par l'(les) auteur(s) |
---|