Distributional Reinforcement Learning with Quantile Regression

Understand

In reinforcement learning an agent interacts with the environment by taking actions and observing the next state and reward.

• When sampled probabilistically, these state transitions, rewards, and actions can all induce randomness in the observed long-term return.
• Traditionally, reinforcement learning algorithms average over this randomness to estimate the value function.
• In this paper, we build on recent work advocating a distributional approach to reinforcement learning in which the distribution over returns is modeled explicitly instead of only estimating the mean.