Why don't today's deep nets overfit to their training data?

Research output: Contribution to journal › Comment/debate › peer-review

Original language	English (US)
Pages (from-to)	106
Number of pages	1
Journal	Communications of the ACM
Volume	64
Issue number	3
DOIs	https://doi.org/10.1145/3446773
State	Published - Mar 2021

All Science Journal Classification (ASJC) codes

General Computer Science

Access to Document

10.1145/3446773

Other files and links

Cite this

@article{08c567086a314d10bb0989e5805c8a6c,

title = "Why don't today's deep nets overfit to their training data?",

author = "Sanjeev Arora",

year = "2021",

month = mar,

doi = "10.1145/3446773",

language = "English (US)",

volume = "64",

pages = "106",

journal = "Communications of the ACM",

issn = "0001-0782",

publisher = "Association for Computing Machinery (ACM)",

number = "3",

}

TY - JOUR

T1 - Why don't today's deep nets overfit to their training data?

AU - Arora, Sanjeev

PY - 2021/3

Y1 - 2021/3

UR - http://www.scopus.com/inward/record.url?scp=85101525291&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85101525291&partnerID=8YFLogxK

U2 - 10.1145/3446773

DO - 10.1145/3446773

M3 - Comment/debate

AN - SCOPUS:85101525291

SN - 0001-0782

VL - 64

SP - 106

JO - Communications of the ACM

JF - Communications of the ACM

IS - 3

ER -