Faster Gradient-TD Algorithms

By Leah Hackman. 2013

21 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Control and Systems Engineering

How researchers see this paper

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

In Aerospace Engineering

1. Martha White
2. A. Mahmood
3. Adam White
4. Richard S. Sutton
5. Andrew Patterson
6. Sina Ghiassian

In Computer Vision and Pattern Recognition

1. Martha White
2. Andrew Patterson
3. Richard S. Sutton
4. A. Mahmood
5. Adam White
6. Sina Ghiassian
7. Giacomo Spigler

In Control and Systems Engineering

1. A. Mahmood
2. Richard S. Sutton
3. Martha White
4. Andrew Patterson
5. Adam White
6. Sina Ghiassian

Citations in the corpus

• An Emphatic Approach to the Problem of Off-policy Temporal-Difference Learning (2015, methodology): “Perhaps the most successful approach to date is the gradient-TD approach (e.g., Maei 2011, Sutton et al. 2009, Maei et al. 2010), including hybrid methods such as HTD (Hackman 2012).”
• Investigating Practical Linear Temporal Difference Learning (2016, methodology): “Later, Hackman[6] produced a hybrid version of the GQ(0) algorithm, estimating state-action value functions rather than state-value functions as we do here.”
• Online Off-policy Prediction (2018, methodology): “Results in on-policy Random MDPs (Mahadevan et al., 2014) and an energy domain (Liu et al., 2016) contradict later studies on Random MDPs (White & White, 2016) that show that conventional Gradient-TD methods perform best in both on and off-policy settings.”