… Here we present an IRL algorithm that considers the case … in nature and the decision process
is decentralized such that … We briefly summarize the work done by Ng and Russell [2000]…

Inverse Reinforcement Learning (IRL) describes the problem of learning an unknown reward
function of a Markov Decision Process (MDP) from observed behavior of an agent. Since …

… [8] improve upon the limitations of Ng and Russell’s[5] … will is formulated using a Markov
Decision Process (MDP). The goal … Q-learning is a reinforcement learning algorithm that takes a …

… maximum entropy inverse reinforcement learning models [39] … Next, we describe in
detail our algorithm for obtaining goal … R(st ) = θ � ϕ(st ), Abbeel & Ng [1] propose the apprenticeship …

… inverse reinforcement learning proposed by Abbeel and Ng [… -world domains showing that
the semi-supervised algorithm … 2000] is to learn a good behavior by observing the behavior …

… We detail a novel approach that uses inverse reinforcement learning to learn a set … , we use
the linear IRL algorithm as described in Section . For Pac-Man, observe that … Ng and Stuart J. …

… assumption that people do inverse reinforcement learning to infer an … Involving a human
user in the evaluation process can help … We modified the algorithm to extract a fixed budget by …

… control tasks over two RL algorithms, shows improvements over … This human-intensive
process raises questions: a) Can we … We train up to ng = 1000 agents parallelized across nmc = …

… shown that KLregularized reinforcement learning from expert … unsolved by standard deep
reinforcement learning algorithms. … model classes for deep reinforcement learning algorithms, …

… work aims at improving this algorithm relative to the training … 2018) to find the optimal policy
for the Markov decision process … Deepmimic: Exampleguided deep reinforcement learning of …