Learning Priors for Bayesian Computations in the Nervous System

Max Berniker, Martin Voss, Konrad Kording · 2010

Read Learning Priors for Bayesian Computations in the Nervous System by Max Berniker, Martin Voss, Konrad Kording (2010) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0012686.txt.

Description

Berniker, Voss and Kording modelled how the brain mixes fresh senses with past learning. People blend what they see and hear now with expectations built over a lifetime. Their 2010 paper studies priors, the brain's stored guesses, and how a learner should set them.

They show how good priors can be learned rather than fixed in advance. Getting this right matters for theories of perception and for building machines that learn like brains. The work links lab findings on human judgement to a clean learning rule. The model also predicts when strong priors help and when they mislead.

Catalog record

CREATORMax Berniker, Martin Voss, Konrad Kording
YEAR2010
COLLECTIONPAPERS
FORMATTXT
FILE SIZE52 KB
LICENSECreative Commons Attribution License
RIGHTSCC BY
SOURCEPLOS ONE

Source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0012686

Browse the full library · About the Tower of Babel