Tractable Objectives for Robust Policy Optimization

By Katherine Chen, Michael H. Bowling. 2012

33 citations recorded by the source corpus.

Subfields: Reinforcement Learning, Optimization & Training

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Reinforcement Learning

1. Neil Burch
2. Dustin Morrill
3. Viliam Lisý
4. Yuxi Li
5. Michael Bradley Johanson
6. Michael H. Bowling
7. Nolan Bard
8. Martin Schmid

In Optimization & Training

1. Nick Hawes
2. Kevin Waugh
3. Michael Bradley Johanson
4. Michael H. Bowling
5. Viliam Lisý
6. Dustin Morrill
7. Neil Burch
8. Martin Schmid

Citations in the corpus

• DeepStack: Expert-level artificial intelligence in heads-up no-limit poker (2017, background): “The abstraction paradigm for handling imperfect information has shown promise in applications like defending strategic resources (42) and robust decision making as needed for medical treatment recommendations (43).”
• Convex Optimization: Algorithms and Complexity (2015, unknown): “Chen and Bowling [Chen and Bowling, 2012] introduce the k-of-N family of percentile measures, which admits an efficient solution under a general uncertainty structure.”
• Risk-Averse Bayes-Adaptive Reinforcement Learning (2021, background): “One approach is to find a policy which optimises the expected value in the worst k% of parameter realisations [5, 11].”