Skip to main content

Jan 29, 2024

Efficient and robust reversal learning

Date: January 29, 2024 | 11:30 am – 12:30 pm
Speaker: Angelika Steger, ETH
Location: Raiffeisen Lecture Hall
Language: English

image here

 

One of the defining characteristics of biological intelligence is its rapid adaptability to new environments, a trait explored in cognitive science through reversal learning tasks.  In this talk we present a novel approach  for the multi-armed bandit problem, the machine learning analogue of the reversal learning paradigm, that departs from the traditional method of estimating values. Instead, our method relies on continuously holding a belief about the best current option, coupled with a verification mechanism that periodically assesses this belief's validity. We demonstrate that our approach not only theoretically outperforms the best known algorithm but also explains well measured data across various experimental settings. 

More Information:

Date:
January 29, 2024
11:30 am – 12:30 pm

Speaker:
Angelika Steger, ETH

Location:
Raiffeisen Lecture Hall

Language:
English

Contact:

Arinya Eller

Email:
arinya.eller@ist.ac.at

Share

facebook share icon
twitter share icon


sidebar arrow up
Back to Top