Probabilistic Tree Search

01/06/2020 to 31/05/2023

Tree search is a classic computational task and yet still relevant for machine learning applications. Phrased as (sequential) active and reinforcement learning, it is arguably the most extreme case of “small-data AI”, since it requires learning in an exponentially large decision space from linearly limited data. This research project will develop a probabilistic formalism for tree search. It will be able to leverage structure of the search domain, and use probabilistic decision theory to locally improve search efficiency. Microsoft Research has agreed to fund the project through their open grant program, which selects based on scientific merit. The contents of the project were actively developed and proposed by Chair for the Methods of Machine Learning in Tübingen. There are no obligations by the research team to produce economic value for Microsoft.