Kernel-Based Reinforcement Learning

By Dirk Ormoneit, Śaunak Sen. 2004

139 citations recorded by the source corpus.

Subfields: Molecular Biology, Reinforcement Learning, Statistical Learning Theory

How researchers see this paper

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

In Molecular Biology

1. Shamim Nemati
2. Peter Stone
3. Richard E. Turner
4. Adrian Weller
5. R´emi Munos
6. Nicholas K. Jong
7. Krzysztof Marcin Choromanski
8. Chao Yu

In Reinforcement Learning

1. Sergey Levine
2. Csaba Szepesvari
3. R´emi Munos
4. Nan Jiang
5. Peter Stone
6. Amir-massoud Farahmand
7. Marc G. Bellemare
8. Pieter Abbeel

In Statistical Learning Theory

1. Csaba Szepesvari
2. R´emi Munos
3. András Antos
4. Vikas Sindhwani
5. Peter Stone
6. Nan Jiang
7. Adrian Weller
8. Krzysztof Marcin Choromanski

Citations in the corpus

• Reinforcement Learning in Healthcare: A Survey (2019, background): “ced Treatment Alternatives to Relieve Depression (STAR*D) trial [164], which is a sequenced four-stage randomized clinical trial of patients with MDD, Pineau et al. [165] first applied Kernelbased BRL [166] for constructing useful DTRs for patients with MDD. Other work tries to address the problem of”
• Model-based function approximation in reinforcement learning (2007, methodology): “This version of AMBI can also be construed as an online implementation of Kernel-Based Reinforcement Learning [12] (KBRL), discussed in Section 6, combined with the exploration mechanism of Prioritized Sweeping.”
• The Geometry of Random Features (2018, methodology): “Kernel methods are a central tool in machine learning, with many applications including classification (SVMs, Cortes and Vapnik, 1995), regression (kernel ridge regression), Gaussian processes (Rasmussen and Williams, 2005), principal component analysis, novelty detection, bioinformatics (graph kern”
• Learning near-optimal policies with Bellman-residual minimization based fitted policy iteration and a single sample path (2006, background): “Another interesting theoretical development concerning off-policy control learning with value-function approximation is the paper by Ormoneit and Sen (2002) who considered kernel-regression in conjunction with approximate value iteration over action-value functions and obtained asymptotic rates on”
• Interpretable Policies for Reinforcement Learning by Genetic Programming (2017, background): “Research from the past two decades (Gordon, 1995; Ormoneit and Sen, 2002; Lagoudakis and Parr, 2003; Ernst, Geurts, Wehenkel, and Littman, 2005) suggests that such batch RL algorithms satisfy real-world system requirements, particularly when involving neural networks (NNs) modeling either the…”