Skip to Content
You are currently on the new version of our website. Access the old version .
AlgorithmsAlgorithms
  • Article
  • Open Access

27 July 2021

23 Pages

Synthetic Experiences for Accelerating DQN Performance in Discrete Non-Deterministic Environments †

,
and
1
Organic Computing Group, University of Augsburg, 86159 Augsburg, Germany
2
Artificial Intelligence in Agricultural Engineering, University of Hohenheim, 70599 Stuttgart, Germany
*
Author to whom correspondence should be addressed.
†
This paper is an extended version of our paper published in 12th International Conference on Neural Computation Theory and Applications, Budapest, Hungary, 2–4 November 2020.
This article belongs to the Special Issue Algorithmic Aspects of Neural Networks

Abstract

State-of-the-art Deep Reinforcement Learning Algorithms such as DQN and DDPG use the concept of a replay buffer called Experience Replay. The default usage contains only the experiences that have been gathered over the runtime. We propose a method called Interpolated Experience Replay that uses stored (real) transitions to create synthetic ones to assist the learner. In this first approach to this field, we limit ourselves to discrete and non-deterministic environments and use a simple equally weighted average of the reward in combination with observed follow-up states. We could demonstrate a significantly improved overall mean average in comparison to a DQN network with vanilla Experience Replay on the discrete and non-deterministic FrozenLake8x8-v0 environment.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.