An algebraic approach to abstraction in reinforcement learning

By Balaraman Ravindran, Andrew Barto. 2004

128 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

Information-Theoretic Considerations in Batch Reinforcement Learning (2019, methodology): “…space of piece-wise constant functions under a partition induced by state abstraction φ , the notion of completeness (Assumption 3, F = G ) is exactly equivalent to a long-studied type of abstractions, known as bisimulation (Whitt, 1978; Even-Dar & Mansour, 2003; Ravindran, 2004; Li et al., 2006).”
Transfer Learning for Reinforcement Learning Domains: A Survey (2009, background): “A second example would be to start with a simple source task and change it overtime by modifying the transition function (Selfridge et al., 1985) and start state (Asada et al.,1994), while learning options (Ravindran and Barto, 2003b), until a difficult target task is learn d.”
An Introduction to Deep Reinforcement Learning (2018, unknown): “The idea of selecting the right features for the task at hand is key in the whole field of machine learning and also highly prevalent in reinforcement learning (see e.g., Munos and Moore, 2002; Ravindran and Barto, 2004; Leffler et al., 2007; Kroon and Whiteson, 2009; Dinculescu and Precup, 2010; Li”
Policy Error Bounds for Model-Based Reinforcement Learning with Factored Linear Models (2016, background): “If R is a point-evaluator, we recover MDP homomorphisms per se (Ravindran, 2004), and the more general case of R stochastic yields soft MRP homomorphisms (Sorg and Singh, 2009).”
Deictic Image Maps: An Abstraction For Learning Pose Invariant Manipulation Policies (2018, methodology): “The analysis section of this paper leverages the MDP Homomorphism framework introduced by Ravindran and Barto (Ravindran 2004; Ravindran and Barto 2003).”
Transfer of Deep Reactive Policies for MDP Planning (2018, background): “, consider a navigation problem over a square grid [Ravindran, 2004], with the goal at top-right corner.”
Abstraction Selection in Model-based Reinforcement Learning (2015, background): “When the quality of the abstraction is known, the theory of approximate homomorphisms in MDPs bounds the loss of the certainty equivalence policy (Even-Dar & Mansour, 2003; Ravindran, 2004).”