Deep Q-Learning For Directed Acyclic Graph Generation

Abstract We present a method to generate directed acyclic graphs (DAGs) using deep reinforcement learning, specifically deep Q-learning. Generating graphs with specified structures is an important and challenging task in various application fields, however most current graph generation methods produce graphs with undirected edges. We demonstrate that this method is capable of generating DAGs with topology and node types satisfying specified criteria in highly sparse reward environments. In ongoing work, we apply this technique to dynamically compose ensembles of neural networks in a coalition setting.
  • Laura D 'Arcy (Cardiff)
  • Padraig Corcoran (Cardiff)
  • Alun Preece (Cardiff)
  • Krishna Kesari (Purdue)
  • Raghu Ganti (IBM US)
Date Sep-2019
Venue Annual Fall Meeting of the DAIS ITA, 2019