Temporal Abstraction in Temporal-difference Networks

52 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.

DEVELOPING A PREDICTIVE APPROACH TO KNOWLEDGE (2015, background): “…(PSRs) (Littman et al., 2002; Singh et al., 2004; Bowling et al., 2006; Wingate & Singh, 2008), temporal difference networks (TD-nets) (Sutton & Tanner, 2005; Sutton et al., 2006), and prediction profiles (Talvitie & Singh, 2011) have also been limited to small, discrete simulations.”
A convergent O ( n ) algorithm for off-policy temporal-difference learning with linear function approximation (2008, methodology): “Off-policy methods are also critical to reinforcement-learning-based efforts to model human-level world knowledge and state representations as predictions of option outcomes (e.g., Sutton, Precup & Singh 1999; Sutton, Rafols & Koop 2006 ).”
Incremental Off-policy Reinforcement Learning Algorithms (2017, background): “…incremental off-policy prediction algorithms (Precup et al. 2000), several works acknowledged the importance of adopting off-policy policy evaluation tasks for representing long-term predictions of sensorimotor events and have put this to work (Sutton et al. 1999, Sutton et al. 2006, Rafols 2006).”
Off-policy Learning with Options and Recognizers (2005, background): “In this paper we focus on learning predictions about expected returns, but other predictions can be tackled as well (for instance, predictions of transition models for options (Sutton, Precup & Singh, 1999), or predictions specified by a TD-network (Sutton & Tanner, 2005; Sutton, Rafols & Koop, 200”