Reinforcement learning for the adaptive control of perception and action

How researchers see this paper

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

Deep learning in neural networks: An overview (2014, neutral): “Compare RL-based attention learning without NNs (Whitehead, 1992).”
Deep Networks with Internal Selective Attention through Feedback Connections (2014, neutral): “combined with traditional computer vision primitives [47].”
Learning Without State-Estimation in Partially Observable Markovian Decision Processes (1994, background): “Unfortunately all of the elegant theory of RL islimited to Markovian decision processes (MDPs) (Sut-ton, 1988; Watkins and Dayan, 1992; Dayan, 1992,Jaakkola et al., 1994, Tsitsiklis, to appear).”
Fast Online Q(λ) (1998, background): “Keywords: reinforcement learning, Q-learning, TD(λ), online Q(λ), lazy learning”
Driven by Compression Progress: A Simple Principle Explains Essential Aspects of Subjective Beauty, Novelty, Surprise, Interestingness, Attention, Curiosity, Creativity, Art, Science, Music, Jokes (2008, background): “It will learn to focus its attention [96, 116] and its activel y chosen experiments on things that are currently still incompressible but are expe cted to become compressible / predictable through additional learning.”
Reward Functions for Accelerated Learning (1994, methodology): “Whitehead (1992) eloquently describes why assumptions about world models commonly held in RL do not apply to real–world tasks.”