Picture for Craig Sherstan

Craig Sherstan

Sony AI

Value Function Decomposition for Iterative Design of Reinforcement Learning Agents

Add code
Jun 24, 2022
Figure 1 for Value Function Decomposition for Iterative Design of Reinforcement Learning Agents
Figure 2 for Value Function Decomposition for Iterative Design of Reinforcement Learning Agents
Figure 3 for Value Function Decomposition for Iterative Design of Reinforcement Learning Agents
Figure 4 for Value Function Decomposition for Iterative Design of Reinforcement Learning Agents
Viaarxiv icon

Work in Progress: Temporally Extended Auxiliary Tasks

Add code
Apr 16, 2020
Figure 1 for Work in Progress: Temporally Extended Auxiliary Tasks
Figure 2 for Work in Progress: Temporally Extended Auxiliary Tasks
Figure 3 for Work in Progress: Temporally Extended Auxiliary Tasks
Figure 4 for Work in Progress: Temporally Extended Auxiliary Tasks
Viaarxiv icon

Gamma-Nets: Generalizing Value Estimation over Timescale

Add code
Nov 23, 2019
Figure 1 for Gamma-Nets: Generalizing Value Estimation over Timescale
Figure 2 for Gamma-Nets: Generalizing Value Estimation over Timescale
Figure 3 for Gamma-Nets: Generalizing Value Estimation over Timescale
Figure 4 for Gamma-Nets: Generalizing Value Estimation over Timescale
Viaarxiv icon

Accelerating Learning in Constructive Predictive Frameworks with the Successor Representation

Add code
Mar 23, 2018
Figure 1 for Accelerating Learning in Constructive Predictive Frameworks with the Successor Representation
Figure 2 for Accelerating Learning in Constructive Predictive Frameworks with the Successor Representation
Figure 3 for Accelerating Learning in Constructive Predictive Frameworks with the Successor Representation
Figure 4 for Accelerating Learning in Constructive Predictive Frameworks with the Successor Representation
Viaarxiv icon

Directly Estimating the Variance of the λ-Return Using Temporal-Difference Methods

Add code
Feb 14, 2018
Figure 1 for Directly Estimating the Variance of the λ-Return Using Temporal-Difference Methods
Figure 2 for Directly Estimating the Variance of the λ-Return Using Temporal-Difference Methods
Figure 3 for Directly Estimating the Variance of the λ-Return Using Temporal-Difference Methods
Figure 4 for Directly Estimating the Variance of the λ-Return Using Temporal-Difference Methods
Viaarxiv icon

Communicative Capital for Prosthetic Agents

Add code
Nov 10, 2017
Figure 1 for Communicative Capital for Prosthetic Agents
Figure 2 for Communicative Capital for Prosthetic Agents
Figure 3 for Communicative Capital for Prosthetic Agents
Figure 4 for Communicative Capital for Prosthetic Agents
Viaarxiv icon

Introspective Agents: Confidence Measures for General Value Functions

Add code
Jun 17, 2016
Figure 1 for Introspective Agents: Confidence Measures for General Value Functions
Viaarxiv icon