Symposium Plausible Neural Reinforcement Learning and Control

On September 3, 2015, CWI will host the symposium Plausible Neural Reinforcement Learning and Control.
  • When 03-09-2015 from 11:00 to 16:00 (Europe/Amsterdam / UTC200)
  • Where room L016, at CWI
  • Add event to calendar iCal

On September 3, 2015, CWI will host the symposium Plausible Neural Reinforcement Learning and Control. 

With AI being fueled by advances in artificial neural networks, this symposium hosts a number of renowned speakers who will discuss plausible neural network models and architectures that deal with learning and control. 
 

Program  

13:15h start program

13:30h Peter Dayan "Heuristics of Control: Habitization, Fragmentation, Memorization and pruning

Goal-directed or model-based control faces substantial computational challenges in deep planning problems. One popular solution to this is habitual or model-free control based on state-action values. In a simple planning task, we found evidence for three further heuristics: blinkered pruning of the decision-tree, the storage and direct reproduction of sequences of actions arising from search, and hierarchical decomposition. I will discuss models of these and their interactions. This is work with Quentin Huys in collaboration with Jon Roiser and Sam Gershman.

14:30h break

14:45h Jaldert Rombouts "Biologically plausible learning of working memory tasks"

15:15h Heiko Neumann "Neural reinforcement learning in recurrent networks with distinct excitatory, inhibitory, and modulatory connections"

Recent neurophysiological studies revealed that reward-based learning influences recurrent grouping processes, but it is not well understood how rewards train recurrent neural circuits for perceptual organization. I examine the mechanisms for reward-based learning of new grouping rules, and derive a learning rule that can explain how rewards influence the information flow through a recurrent counter-stream model of cortical visual processing. The model capacity is illustrated with two tasks that have been used to study the neuronal correlates of perceptual organization in early visual cortex, including curve-tracing. 

15:45h break

16:00h Bert Kappen "Learning Optimal Control"

16:30h Davide Zambrano "Continuous-time on-policy neural reinforcement learning of working memory tasks"

17:00h drinks 

You are cordially invited to attend the symposium. Please register without further costs.