The attention-gated reinforcement learning model: performance and predictions.
A problem with many traditional neural network learning algorithms is that they lack a biologically plausible method of assigning credit to the cells in the earlier levels of the network that play a decisive role in the stimulus response mapping. In the proposed Attention Gated Reinforcement Learning Model (AGREL) this so-called credit assignment problem is overcome by the focal influence of attention during learning. Each trial a global reward value (conveyed by neuromodulators) is calculated which serves to increase the likelihood that a successful, and decrease the likelihood that an unsuccessful, behavior be repeated. The activity in the output level of the network is then channeled back through the network via the reciprocal feedback connections, which therefore selectively target the lower level nodes that were important for generating the current output. The learning in the network depends on the interaction of the reward and attention factors, so only the lower level nodes that receive feedback will have their weights modified. This attentional gating of the reward signal offers a biologically realistic solution to the credit assignment problem and in so doing provides a coherent and unifying framework for learning, with attentional selection at its core. Model neurons in AGREL are shown to exhibit similar changes in tuning properties following categorization training, as have been observed in primate cortical neurons. AGREL also generates several predictions concerning the extent of the learning impairments caused by distortions in the reward signal, or the partial removal or loss of the cortical feedback connections.
Conference:
Computational and systems
neuroscience 2009, Salt Lake City, UT, United States, 26 Feb - 3 Mar, 2009.
Presentation Type:
Poster Presentation
Topic:
Poster Presentations
Citation:
(2009). The attention-gated reinforcement learning model: performance and predictions..
Front. Syst. Neurosci.
Conference Abstract:
Computational and systems
neuroscience 2009.
doi: 10.3389/conf.neuro.06.2009.03.010
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
29 Jan 2009;
Published Online:
29 Jan 2009.