Gentle Nearest Neighbors Boosting over Proper Scoring Rules
Tailoring nearest neighbors algorithms to boosting is an important problem. Recent papers study an approach, UNN, which provably minimizes particular convex surrogates under weak assumptions. However, numerical issues make it necessary to experimentally tweak parts of the UNN algorithm, at the possi...
| Veröffentlicht in: | IEEE transactions on pattern analysis and machine intelligence. - 1979. - 37(2015), 1 vom: 01. Jan., Seite 80-93 |
|---|---|
| 1. Verfasser: | |
| Weitere Verfasser: | , , , |
| Format: | Online-Aufsatz |
| Sprache: | English |
| Veröffentlicht: |
2015
|
| Zugriff auf das übergeordnete Werk: | IEEE transactions on pattern analysis and machine intelligence |
| Schlagworte: | Journal Article |
| Zusammenfassung: | Tailoring nearest neighbors algorithms to boosting is an important problem. Recent papers study an approach, UNN, which provably minimizes particular convex surrogates under weak assumptions. However, numerical issues make it necessary to experimentally tweak parts of the UNN algorithm, at the possible expense of the algorithm's convergence and performance. In this paper, we propose a lightweight Newton-Raphson alternative optimizing proper scoring rules from a very broad set, and establish formal convergence rates under the boosting framework that compete with those known for UNN. To the best of our knowledge, no such boosting-compliant convergence rates were previously known in the popular Gentle Adaboost's lineage. We provide experiments on a dozen domains, including Caltech and SUN computer vision databases, comparing our approach to major families including support vector machines, (Ada)boosting and stochastic gradient descent. They support three major conclusions: (i) GNNB significantly outperforms UNN, in terms of convergence rate and quality of the outputs, (ii) GNNB performs on par with or better than computationally intensive large margin approaches, (iii) on large domains that rule out those latter approaches for computational reasons, GNNB provides a simple and competitive contender to stochastic gradient descent. Experiments include a divide-and-conquer improvement of GNNB exploiting the link with proper scoring rules optimization |
|---|---|
| Beschreibung: | Date Completed 24.11.2015 Date Revised 10.09.2015 published: Print Citation Status PubMed-not-MEDLINE |
| ISSN: | 1939-3539 |
| DOI: | 10.1109/TPAMI.2014.2307877 |