English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper addresses a fundamental question: how good are our currentself-supervised visual representation learning algorithms relative to humans?More concretely, how much "human-like" natural visual experience would thesealgorithms need in order to reach human-level performance in a complex,realistic visual object recognition task such as ImageNet? Using a scalingexperiment, here we estimate that the answer is several orders of magnitudelonger than a human lifetime: typically on the order of a million to a billionyears of natural visual experience (depending on the algorithm used). We obtaineven larger estimates for achieving human-level performance in ImageNet-derivedrobustness benchmarks. The exact values of these estimates are sensitive tosome underlying assumptions, however even in the most optimistic scenarios theyremain orders of magnitude larger than a human lifetime. We discuss the maincaveats surrounding our estimates and the implications of these surprisingresults.

How much human-like visual experience do current self-supervised learning algorithms need in order to achieve human-level object recognition?