An Evaluation of Using Deterministic Heuristics to Accelerate Reinforcement Learning

Abstract

Neural networks frequently face long training times based on the corpus of data available to them. Reinforcement learning in particular can take a long time to attain satisfactory performance. Recent efforts to incorporate deterministic logical rules and physical laws into a neural network have met with promising results. From an existing baseline neural network that is designed to learn Pong strictly from pixel representation of the game board, this thesis adds a ball trajectory-based heuristic into the learning process and evaluates its performance. The evaluation initially shows game score improvements, but demonstrates a sharp score degradation after about 25,000 games. Another evaluation shows the heuristic incurs a training time increase of approximately 35%. More work remains for assessing the long-term viability of this approach.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 2018
Accession Number
AD1069772

Entities

People

  • Garret M Walton

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Autonomy
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Artificial Neural Networks
  • Automata Theory
  • California
  • Computer Programming
  • Computer Science
  • Computers
  • Convolutional Neural Networks
  • Data Mining
  • Data Science
  • Data Set
  • Deep Learning
  • Digital Data
  • Information Science
  • Learning
  • Machine Learning
  • Network Science
  • Neural Networks
  • Reinforcement Learning
  • Supervised Machine Learning
  • Test And Evaluation
  • Training
  • Trajectories
  • United States
  • Video Games

Fields of Study

  • Computer science

Readers

  • Computational Modeling and Simulation
  • Neural Network Machine Learning.
  • Operations Research

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks