This repo stores the code used to generate the results presented in the paper "Robust Trading via Adversarial Reinforcement Learning". The main scripts are stored in src, including the actual experiments scripts (bin), environment implementations (env) and RL agent types (agents). These will remain largely unchanged in the future and make use of the rsrl and lfa crates available in Rust.
Due to rapid development, local copies of rsrl, lfa and rstat have been
included in the repo with changes diverging from the main repo. Most of these
have since been integrated into the frameworks but some work is still required.
Once this has been finished, they will be removed from the repo and standard
versioning will resume. At this point further documentation of this repo will
be added for your convenience.