no code implementations • 30 Oct 2019 • Aakash Maroti
This paper proposes a new approach to this $\varepsilon$ decay where the decay is based on feedback from the environment.
Reinforcement Learning (RL)