Advantage Updating Applied to a Differrential Game

By Mance E. Harmon, Leemon C. Baird III, A. Harry Klopf. 1994

Subfields: Reinforcement Learning, Game & Decision Theory

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

In Game & Decision Theory

Citations in the corpus

Taming the Noise in Reinforcement Learning via Soft Updates (2015, background): “In noisy environments, learning A is shown in some examples to be faster than learning Q [23, 24].”
Dueling Network Architectures for Deep Reinforcement Learning (2015, background): “The dueling architecture represents both the value V ( s ) and advantage A ( s, a ) functions with a single deep model whose output combines the two to produce a state-action value Q ( s, a ) .”
Reinforcement Learning Through Gradient Descent (1999, methodology): “5.1 Introduction In work done before the development of the residual algorithms, an algorithm called advantage updating (Harmon, Baird, and Klopf, 1995) was proposed that seemed preferable to Q-learning for continuous-time systems.”
Approximate policy iteration: a survey and some new methods (2011, unknown): “[91,92] and Bertsekas [93]).”
Robust Reinforcement Learning (2005, unknown): “The min-max RL was also applied to a problem in which an airplane tries to avoid a missile and the missile tries to catch the airplane ( Harmon, Baird, & Klopf, 1995 ).”
A Counterexample to Temporal Differences Learning (1995, background): “equation 3.6) have been independently developed by Baird and are briefly described in Baird (1993) and Harmon et al. (1994) (this was pointed out by one of the reviewers).”
Reinforcement Learning: A Tutorial. (1997, unknown): “A more detailed discussion of this alternative can be found in Harmon, Baird, and Klopf (1994), and Littman(1996).”