Learning decisions: robustness, uncertainty, and approximation

By J. Bagnell, Jeff Schneider, Andrew W. Moore, Alfred A. Rizzi. 2004

34 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Reinforcement learning in robotics: A survey (2013, methodology): “Random Numbers Bagnell and Schneider (2001); Bagnell (2004); Ko et al.”
Non-parametric policy gradients: a unified treatment of propositional and relational domains (2008, background): “Most closely related to NPPG is the work of Bagnell and Schneider (2003), see also (Bagnell, 2004).”
Learning to search: structured prediction techniques for imitation learning (2009, methodology): “This section makes extensive use of RKHS functional gradients to aid in deriving the least-squares technique for projecting a Euclidean functional gradient onto a space of functions; we do not review RKHS functional gradients in detail here, but an introduction to these concepts can be found in (Bag”
A comprehensive survey on safe reinforcement learning (2015, background): “…the risk is one of several imports in model-based reinforcement learning solutions , particularly ones in which failure has important consequences (Bagnell and Schneider, 2008), as a learned model invariably has certain inaccuracies, due to insufficient or noise training data (Bagnell, 2004).”
Nonparametric Return Distribution Approximation for Reinforcement Learning (2010, background): “It is known that these risk-taking and risk-aversion strategies lead to the exploration and exploitation (robust) behaviors, respectively (Bagnell, 2004).”
Policy gradient methods (2010, background): “[1] James Andrew Bagnell.”
Parametric Return Density Estimation for Reinforcement Learning (2010, background): “Recently, such risk-averse scenarios have also been considered in RL as risk-sensitive RL (Heger, 1994; Mihatsch and Neuneier, 2002; Bagnell, 2004).”
CHOMP: Covariant Hamiltonian optimization for motion planning (2013, background): “, 2009, 2010), and robust and stochastic control (Bagnell, 2004; Todorov, 2006; Theodorou et al., 2010).”