English
 
User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Engineering a Less Artificial Intelligence

Sinz, F., Pitkow, X., Reimer, J., Bethge, M., & Tolias, A. (2019). Engineering a Less Artificial Intelligence. Neuron, 103(6), 967-979. doi:10.1016/j.neuron.2019.08.034.

Item is

Basic

show hide
Item Permalink: http://hdl.handle.net/21.11116/0000-0004-C2A5-6 Version Permalink: http://hdl.handle.net/21.11116/0000-0004-C2A6-5
Genre: Journal Article

Files

show Files

Locators

show

Creators

show
hide
 Creators:
Sinz, FH, Author              
Pitkow, X, Author
Reimer, J, Author
Bethge, M1, 2, Author              
Tolias, AS, Author              
Affiliations:
1Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497794              
2Research Group Computational Vision and Neuroscience, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497805              

Content

show
hide
Free keywords: -
 Abstract: Despite enormous progress in machine learning, artificial neural networks still lag behind brains in their ability to generalize to new situations. Given identical training data, differences in generalization are caused by many defining features of a learning algorithm, such as network architecture and learning rule. Their joint effect, called “inductive bias,” determines how well any learning algorithm—or brain—generalizes: robust generalization needs good inductive biases. Artificial networks use rather nonspecific biases and often latch onto patterns that are only informative about the statistics of the training data but may not generalize to different scenarios. Brains, on the other hand, generalize across comparatively drastic changes in the sensory input all the time. We highlight some shortcomings of state-of-the-art learning algorithms compared to biological brains and discuss several ideas about how neuroscience can guide the quest for better inductive biases by providing useful constraints on representations and network architecture.

Details

show
hide
Language(s):
 Dates: 2019-09
 Publication Status: Published in print
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Method: -
 Identifiers: DOI: 10.1016/j.neuron.2019.08.034
 Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show
hide
Title: Neuron
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: Cambridge, Mass. : Cell Press
Pages: - Volume / Issue: 103 (6) Sequence Number: - Start / End Page: 967 - 979 Identifier: ISSN: 0896-6273
CoNE: https://pure.mpg.de/cone/journals/resource/954925560565