The Strength of Weak Learnability

Schapire, R. E. (1990). The Strength of Weak Learnability. Machine Learning, 5(2), 197-227.

Type

Journal article

Intellectual base

Primary research

Year

1990

Volume

5

Issue

2

Pages

197-227

DOI

10.1007/bf00116037

DOI registrant

Springer Nature

Link

https://doi.org/10.1007/BF00116037

Cited from

pharmacology_toxicology

Cited by

3 citations across 3 artifacts.

Each citation links to the sentence it supports in the citing article.

Primes

• Potentiation
  • The construct manages the complexity of dynamic response by providing a named category for history-dependent sensitization—structurally parallel to tolerance but mechanistically and directionally distinct—that enables separate characterization of sensitizing mechanism, temporal course, dose-response dynamics, and management strategy, in much the way Schapire (1990) demonstrated that sequential weak-learner amplification (boosting) can be reasoned about as a binary directional decision before mechanistic elaboration.

    This sourceFoundational boosting result: sequential weak-learner amplification yields disproportionate aggregate response, framed via a binary directional decision (boost vs. not) before mechanistic elaboration of the amplification scheme.

    Supported in partVerified against the work's full text

    Supports only the founding boosting result (weak learner converted to arbitrarily high accuracy); it does not frame boosting as a binary directional decision before mechanistic elaboration.

    “A method is described for converting a weak learning algorithm into one that achieves arbitrarily high accuracy.”

    Read by an automated reader; not a human review. How support was checked

Domain-specific

• AdaBoost
  • The theoretical weak-learning condition means performance with an edge under the distributions presented to it, not small model size, shallow trees, or poor quality in ordinary language.
• Boosting
  • Schapire proved the equivalence of weak and strong learnability and gave a boosting construction.

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 3 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:53a67a6e7fd4 · see in the full table