The study of random instances of NP complete and coNP complete problems has had much impact on our understanding of the nature of hard problems. In this work, we initiate an effor...
Recent research has shown the benefit of framing problems of imitation learning as solutions to Markov Decision Problems. This approach reduces learning to the problem of recoveri...
Brian Ziebart, Andrew L. Maas, J. Andrew Bagnell, ...
In reinforcement learning, least-squares temporal difference methods (e.g., LSTD and LSPI) are effective, data-efficient techniques for policy evaluation and control with linear v...
Michael H. Bowling, Alborz Geramifard, David Winga...
Scoring rules are a broad and concisely-representable class of voting rules which includes, for example, Plurality and Borda. Our main result asserts that the class of scoring rul...
Ariel D. Procaccia, Aviv Zohar, Jeffrey S. Rosensc...
The goal of Reinforcement learning (RL) is to maximize reward (minimize cost) in a Markov decision process (MDP) without knowing the underlying model a priori. RL algorithms tend ...