Recent advances in neural networks have resulted in reliable predictions of observations in Atari games. The quality of predictions can be improved by addition of a generative adversarial approach to the training. This research investigates this development and the benefits of unsupervised predictive learning for reinforcement learning agents.
|Number of pages||2|
|Publication status||Published - 7 Jun 2017|
|Event||2017 EPSRC CDT Student Conference – Oxford, Bristol and Edinburgh - Oxford, United Kingdom|
Duration: 7 Jun 2017 → 7 Jun 2017
|Conference||2017 EPSRC CDT Student Conference – Oxford, Bristol and Edinburgh|
|Period||7/06/17 → 7/06/17|