Reconciling Modern Machine-Learning Practice and the Classical Bias–Variance Trade-off.¶
Belkin, M., Hsu, D., Ma, S., & Mandal, S. (2019). Reconciling Modern Machine-Learning Practice and the Classical Bias–Variance Trade-off. Proceedings of the National Academy of Sciences, 116(32), 15849-15854.
Cited by¶
2 citations across 2 artifacts.
Each citation links to the sentence it supports in the citing article.
Primes¶
- Grain of Analysis
- Categorical classification and neural architecture: class boundaries and parameter counts set grain; over-stratification destroys statistical leverage, under-stratification destroys substantive resolution, and the double-descent literature concerns non-monotone grain–performance relations.
This sourceDocuments the double-descent curve, a non-monotone relation between model capacity (grain) and generalization performance.
Supported in partVerified against the work's full text
Documents double descent, a non-monotone relation between model capacity (parameter count) and test performance, but not class boundaries or stratification.
“increasing model capacity beyond the point of inter- polation results in improved performance”
- Categorical classification and neural architecture: class boundaries and parameter counts set grain; over-stratification destroys statistical leverage, under-stratification destroys substantive resolution, and the double-descent literature concerns non-monotone grain–performance relations.
- Overfitting
- Listed in the references but not attached to a specific claim.
Verification¶
Does it exist? Confirmed. This work's DOI resolves to a registered record, which fixes its identity. That is all it fixes.
Does it back the claim? Read against the text for 1 of 2 citations: 1 supported in part. Each verdict is shown under its citation below, with what in the work backs the sentence.
Support is checked per citation rather than per work — the same source can be cited soundly in one article and wrongly in another. Per-citation recording began recently, so a citation with no recorded check is a gap in the record rather than evidence it went unchecked.
See how references were verified.
Registry ID ref:08119f7f4e3a · see in the full table