qarl

Python library for Quantile Assessment using Reinforcement Learning (RL).

key features

• estimation under poor supervising (guesses of the agent are corrected only by boolean feedback "less" or "greater")
• openendedness (estimation of non-stationary distributions)
• estimation of distribution of random variable
• estimation of distribution of non-stationary stochastic variable