Fitted Q-Learning for Relational Domains

no code implementations • 10 Jun 2020 • Srijita Das, Sriraam Natarajan, Kaushik Roy, Ronald Parr, Kristian Kersting

We consider the problem of Approximate Dynamic Programming in relational domains.

Q-Learning

On the Existence of Simpler Machine Learning Models

no code implementations • 5 Aug 2019 • Lesia Semenova, Cynthia Rudin, Ronald Parr

We hypothesize that there is an important reason that simple-yet-accurate models often do exist.

BIG-bench Machine Learning Fairness +2

Proceedings of the Twenty-Third Conference on Uncertainty in Artificial Intelligence (2007)

no code implementations • 25 Aug 2012 • Ronald Parr, Linda S. van der Gaag

This is the Proceedings of the Twenty-Third Conference on Uncertainty in Artificial Intelligence, which was held in Vancouver, British Columbia, July 19 - 22 2007.

Linear Complementarity for Regularized Policy Evaluation and Improvement

no code implementations • NeurIPS 2010 • Jeffrey Johns, Christopher Painter-Wakefield, Ronald Parr

We demonstrate that warm starts, as well as the efficiency of LCP solvers, can speed up policy iteration.

feature selection Reinforcement Learning (RL)

Least-Squares Policy Iteration

1 code implementation • 4 Dec 2003 • Michail G. Lagoudakis, Ronald Parr

We propose a new approach to reinforcement learning for control problems which combines value-function approximation with linear architectures and approximate policy iteration.

Q-Learning reinforcement-learning +1