Adaptive Machine Learning of Interaction Sequences (AMALIS) (Completed)
2008–2011
This project explored how systems can learn multivariate sequences of interaction data in highly dynamic environments. We developed models for unsupervised and reinforcement learning of hierarchical structure from sequential data (Ordered Means Models), which not only afford the analysis of sequences but also generation of learned patterns. This was demonstrated, e.g., in enabling the agent VINCE to play, and always win the rock-paper-scissors game.
