| 日時(Date) | This talk has been postponed. The resucheduled date will be announced later.Wed. Feb. 2nd, 2022, 3rd period (13:30--15:00) |
|
|---|---|---|
| 場所(Location) | online | |
| 司会(Chair) | 福嶋 | |
| 講演者(Presenter) | Dr. Aurelio Cortese (ATR) / Aurelio Cortese（ATR 主任研究員） | |
| 題目(Title) | The interplay between cognitive functions and reinforcement learning | |
| 概要(Abstract) | A central issue in reinforcement learning (RL) is the ‘curse-of-dimensionality’, which is a problem arising when the degrees-of-freedom are much larger than the number of training samples. In such circumstances, the learning process becomes too slow to be feasible for any biological organism. In the brain, higher cognitive functions (such as attention, memory, but also metacognition) may provide a solution by generating low dimensional mental representations that are suitable for RL to efficiently operate. In this talk I will first introduce human reinforcement learning and the idea that cognitive functions can “augment” learning algorithms. I will then discuss a series of studies from my group that used functional magnetic resonance imaging (fMRI) and computational modelling to investigate the neuro-computational basis of efficient RL. We found that people can learn remarkably complex task structures non-consciously, but also that - intriguingly - metacognition appears tightly coupled to this learning ability. Furthermore, when people use an explicit (conscius) policy to select relevant information, learning is accelerated by abstractions. At the neural level, prefrontal cortex subregions are differentially involved in separate aspects of learning: dorsolateral prefrontal cortex pairs with metacognitive processes, while ventromedial prefrontal cortex with valuation and abstraction. I will discuss the implications of these findings, in particular new questions on the function of metacognition in adaptive behaviour and the link with abstraction. | |
| 講演言語(Language) | English | |
| 講演者紹介(Introduction of Lecturer) | Aurelio Cortese, phd Chief researcher, PI for the ERATO brain-AI project Computational Neuroscience Labs ATR | Advanced Telecommunications Research Institute International, Kyoto - Japan google scholar |