Shaping Mario with Human Advice

In this demonstration, we allow humans to interactively advise a Mario agent during learning, and observe the resulting changes in performance, as compared to its unadvised counterpart. We do this via a novel potential-based reward shaping framework, capable for the first time of handling the scenario of online feedback. Categories and Subject Descriptors I.2.6 [Artificial Intelligence]: Learning Keywords potential-based reward shaping; human advice

• Type: Journal
• Year: 2015
• Where: ATAL
• Authors: Anna Harutyunyan, Tim Brys, Peter Vrancx, Ann Nowé
• Keywords: ATAL 2015, Intelligent Agents