Experiments Focused on Exploration in Deep Reinforcement Learning

Abstract

Automation with deep neural networks is popular topic. Personal artificial assistance, self driving cars and recommendations algorithms, they all use reinforcement learning. However, there are many question about this technology still unanswered. Can the learning process be accelerated, why sometimes the automatons are learning wrong behavior, is it possible for machines teach each others. This papers try to answer those questions with experiments.

Topics & Concepts