Generalized Value Functions for Large Action Sets

51 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.

Discrete Sequential Prediction of Continuous Actions for Deep RL (2017, background): “By laying out actions on a hypercube, [33] are able to perform a binary search over actions resulting in a logarithmic search for the optimal action.”
Non-Parametric Approximate Linear Programming for MDPs (2011, background): “Finally, an advantage of NP-ALP is that it incorporates large and infinite action spaces very naturally, while traditional ALP methods require significant extensions for large action spaces (Pazis and Parr, 2011).”
Learn What Not to Learn: Action Elimination with Deep Reinforcement Learning (2018, background): “Most of the prior work concentrated on factorizing the action space into binary subspaces (Pazis and Parr, 2011; Dulac-Arnold et al., 2012; Lagoudakis and Parr, 2003).”
Deep Reinforcement Learning in Large Discrete Action Spaces (2015, background): “Generalized value functions were proposed in the form of H-value functions (Pazis & Parr, 2011), which allow for a policy to evaluate log( A | ) binary decisions to act.”