How much human-like visual experience do current self-supervised learning algorithms need in order to achieve human-level object recognition?. (arXiv:2109.11523v3 [cs.CV] UPDATED)

This paper addresses a fundamental question: how good are our current
self-supervised visual representation learning algorithms relative to humans?
More concretely, how much "human-like" natural visual experience would these
algorithms need in order to reach human-level performance in a complex,
realistic visual object recognition task such as ImageNet? Using a scaling
experiment, here we estimate that the answer is several orders of magnitude
longer than a human lifetime: typically on the order of a million to a billion
years of …

algorithms arxiv cv experience human human-like learning self-supervised learning supervised learning