Generalized Prioritized Sweeping

By David Andre, Nir Friedman, Ronald E. Parr. 1997

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

Citations in the corpus

Prioritized Experience Replay (2015, background): “Prioritized sweeping (Moore & Atkeson, 1993; Andre et al., 1998) selects which state to update next, prioritized according to the change in value, if that update was executed.”
A survey and critique of multiagent deep reinforcement learning (2018, methodology): “which was in turn inspired by model-based RL algorithms like prioritized sweeping [112, 113].”
Model based Bayesian Exploration (1999, background): “Generalized prioritized sweeping ( Andre, Friedman & Parr 1997 ) allows us to extend prioritized sweeping to these approximate settings.”
Automated Planning and Acting (2016, unknown): “[20], with a focus mechanism in Ferguson”
Perseus: Randomized Point-based Value Iteration for POMDPs (2005, background): “This allows an algorithm to focus on backups which may have a high potential impact, as for instance in the prioritized sweeping algo-rithm for solving fully observable MDPs (Moore & Atkeson, 1993; Andre, Friedman, & Parr, 1998).”
Flexible Decomposition Algorithms for Weakly Coupled Markov Decision Problems (1998, background): “One way to view this result is that it enables a form of high-level prioritized sweeping (Moore & Atkeson, 1993; Andre, Friedman, & Parr, 1997 ).”
Structure-Adaptable Digital Neural Networks (1999, methodology): “A number of new algorithms exploiting this idea and optimizing the use of the internal model of the world have been proposed [11, 142, 151], but we have not analyzed if they are suitable for a digital implementation.”
Regret Minimization Experience Replay in Off-Policy Reinforcement Learning (2021, background): “In model-based planning, Prioritized Sweeping [13, 14, 15] selects the next state updates according to changes in value.”
Experience Replay with Likelihood-free Importance Weights (2020, background): “In model-based planning, Prioritized Sweeping [25, 1, 41] selects the next state updates according to changes in value.”
A Bayesian Framework for Reinforcement Learning (2000, methodology): “The DP solution method is made efficient by avoiding unnecessary iteration of the DP “backup” operation which propagates estimates of Q through the MDP (equation 1). This is achieved by the prioritized sweeping algorithm (Moore & Atkeson, 1993; Andre, Friedman & Parr, 1998 ) which updates only the ”
Hierarchical control and learning for markov decision processes (1998, background): “One way to view this result is that it enables a form of high-level prioritized sweeping (Moore & Atkeson, 1993; Andre et al., 1997).”
A unifying framework for computational reinforcement learning theory (2009, methodology): “The GenPS algorithm [Andre et al. 1998] updates H(s) so that it is always the absolute Bellman error in state s.”