Inverse reinforcement learning in continuous time and space
arXiv:1801.0766311 citationsh-index: 27
Originality Synthesis-oriented
AI Analysis
Provides a data-driven approach to inferring cost functions in continuous time and space for linear systems, addressing a gap in online inverse RL for this setting.
Developed an online inverse reinforcement learning method for linear systems that estimates the cost function from input-output data, achieving estimation up to a multiplicative constant.
This paper develops a data-driven inverse reinforcement learning technique for a class of linear systems to estimate the cost function of an agent online, using input-output measurements. A simultaneous state and parameter estimator is utilized to facilitate output-feedback inverse reinforcement learning, and cost function estimation is achieved up to multiplication by a constant.