Marc VelayCustom Gymnasium Envs and Multi-Input Stable-Baselines3 AgentsCreate a custom Gymnasium environment using Dict spaces and process them with stable-baselines3's MultiInput Policy.·5 min read·Aug 7, 2023----
Marc VelayinTowards Data ScienceEasily Tune RL HyperParameters with Hydra's Optuna sweeperReinforcement Learning agents are very susceptible to their hyperparameters. Easily find the best combo with Hydra and Optuna!·8 min read·Feb 1, 2023----
Marc VelayinTowards Data ScienceReinforcement Learning Intro: Markov Decision ProcessA Markov Decision Process is fundamental in RL. We focus on presenting it as a general mathematical framework and its main difficulties.·9 min read·Aug 16, 2022--1--1
Marc VelayinTowards Data ScienceTarget Networks: Slow and Steady Wins the RaceDive deep into Target Networks and why they have become so popular in state-of-the-art Reinforcement Learning algorithms.·5 min read·Feb 17, 2022--1--1
Marc VelayImplementing Hawkes Processes in Python using tickFew online ressources present how to implement HPs in Python. Here is a tutorial on exactly that, in python using the tick library!·2 min read·Dec 2, 2020----
Marc VelayA brief introduction to Hawkes ProcessesHawkes processes are a type of stochastic processes. These are used to model stochastic — hear random — Point Processes. Hawkes processes…·3 min read·Nov 19, 2020--1--1