Three papers from the Interactive Intelligence group to appear at the NeurIPS 2020 Conference

News - 01 October 2020 - Communication

Frans Oliehoek, whose research focuses on interactive learning and decision making, is the coauthor of three accepted papers:

  1. "Influence Augmented Online Planning for Complex Environments" was co-written with Jinke He and Miguel Suau. It investigates faster online planning by learning more compact simulators.
  2. "MDP Homomorphic Networks: Group Symmetries in Reinforcement Learning" co-written with Elise van der Pol, Daniel E. Worrall, Herke van Hoof and Max Welling explores how symmetries in RL problems can be exploited by neural networks.
  3. "Multi-agent active perception with prediction rewards" co-written with Mikko Lauri investigates the problem of decentralized active perception and how we can use decentralized rewards for correct predictions.

The purpose of the Neural Information Processing Systems annual meeting is to foster the exchange of research on neural information processing systems in their biological, technological, mathematical, and theoretical aspects. The conference is multi-track and interdisciplinary and brings together researchers in machine learning, computational neuroscience, and their applications.