D-CIS Publication Database

Bakker, Bram    

Firstname:Bram 
Surname:Bakker 
Initials:B. 
Email:bram@science.uva.nl 
Institute:University of Amsterdam 
Homepage: 

5 publications

9 Keywords relate to this author

Backpropagation
Behaviours to subgoals
Hierarchical assigment
Hierarchical methods
LSTM model
Multiple qualitatively different state representations
Networked adaptive interactive systems
Reinforcement learning
Traffic light control




Publications as Author



2008

van Seijen, Harm, Bakker, Bram and Kester, Leon, Switching between Different State Representations in Reinforcement Learning, in: Proceedings of the Artificial Intelligence and Applications Conference, 2008.
 

2007

van Seijen, Harm, Bakker, Bram and Kester, Leon, Reinforcement Learning with Multiple, Qualitatively Different State Representations, in: NIPS workshop Hierarchical methods, Vancouver, Canada, 2007.
 
Bakker, Bram, Reinforcement learning by backpropagation through an LSTM model / critic, in: IEEE ADPRL'07, Honolulu, USA, 2007.
 
Bakker, Bram, Hierarchical Assignment of Behaviours to Subgoals, in: NIPS workshop Hierarchical methods, Vancouver, Canada, 2007.
 

2006

Bakker, Bram, Traffic Light Control with Reinforcement Learning Networked Adaptive Interactive Systems, in: Proceedings of COGIS 2006: COGnitive systems with Interactive Sensors (COGIS 2006), 15-17 March 2006, Paris, 2006.