Scaling Up Average Reward Reinforcement Learning by Approximating the Domain Models and the Value Function

By Prasad Tadepalli, DoKyeong Ok. 1996

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 (2019, neutral): “• The calculation of the reward in each task is entirely determined by the objective to achieve and constraints such as differentiability does not apply (Tadepalli and Ok, 1996; Henderson et al., 2018).”
Efficient Solution Algorithms for Factored MDPs (2003, methodology): “Tadepalli and Ok (1996) were the first to apply linear value function approximation to Factored MDPs. Linear value function approximation is a potentially more expressive approximation method because it can assign unique values to every state in an MDP without requiring storage space that is…”
Reinforcement Learning: An Introduction (1998, background): “Tadepalli and Ok (1996) obtained promising results with locally-weighted linear regression to learn a value function for a simulated automatic guided vehicle task.”
Using deep Q-learning to understand the tax evasion behavior of risk-averse firms (2018, methodology): “This approach includes algorithms such as R-learning (Singh, 1994; Tadepalli and Ok, 1996), and Q-Learning (Sutton and Barto, 1998; Tsitsiklis, 1994).”
Hierarchical multi-agent reinforcement learning (2001, methodology): “Previously, Tadepalli and Ok [16] studied a single-agent AGV scheduling task using \ at" average-reward reinforcement learning.”
Accelerating Reinforcement Learning by Composing Solutions of Automatically Identified Subtasks (2011, background): “The rst issueis addressed by Peng (1995) and by Tadepalli and Ok (1996) who use learned instancescombined with linear regression over a set of neighboring points.”
Computing Factored Value Functions for Policies in Structured MDPs (1999, methodology): “This idea is best known under the name value function approximation, which is used frequently in the context or reinforcement learning [Tadepalli and Ok, 1996; Van Roy, 1998].”
Generalized Prioritized Sweeping (1997, methodology): “DBNs have been combined with reinforcement learning before i n [8], where they were used primarily as a means getting better generalization w h le learning.”
Reinforcement learning for factored Markov decision processes (2002, methodology): “Factored models of fully-observable MDPs were learned in [Tadepalli and Ok 1996].”