The Grand Challenge of Predictive Empirical Abstract Knowledge
By Richard S. Sutton. 2009
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- A. Mahmood
- Martha White
- Huizhen Yu
- Richard S. Sutton
- Adam White
- Sergey Levine
- Shangtong Zhang
- Hado van Hasselt
Citations in the corpus
- An Emphatic Approach to the Problem of Off-policy Temporal-Difference Learning (2015, background): “Sutton (2009, 2012) has suggested that parametric TD methods could be key not just to learning about reward, but to the learning of world knowledge generally, and to perceptual learning.”
- GQ(lambda): A general gradient algorithm for temporal-difference prediction learning with eligibility traces (2010, methodology): “Recent theoretical developments in temporal-difference learning combined with mathematical ideas developed for temporally abstract options, known as intra-option learning, can be used to address this challenge (Sutton, 2009).”
- Weak Convergence Properties of Constrained Emphatic Temporal-difference Learning with Constant and Slowly Diminishing Stepsize (2015, background): “In addition, insofar as value functions (with respect to different reward/cost assignments) reflect statistical properties of future outcomes, off-policy learning can be used by an autonomous agent to build an experience-based internal model of the world in artificial intelligence applications [43].”
- Multi-timescale nexting in a reinforcement learning robot (2011, background): “It has even been proposed that all world knowledge can be represented by a sufficiently large and diverse set of predictions (Sutton 2009).”
- Beyond Reward: The Problem of Knowledge and Data (2011, background): “This is the case of greatest interest because in it the system can correct and learn its knowledge autonomously, enabling scaling to very large knowledge bases (see Sutton 2009, 2001).”
- DEVELOPING A PREDICTIVE APPROACH TO KNOWLEDGE (2015, background): “The majority of prior work on predictive knowledge (e.g., Drescher, 1991; Sutton, 2009; Sutton, 2011) has focused on small, discrete simulation worlds, with the notable exception of the work of Boots et al.”