Methods and apparatus for building a support vector machine classifier
US6327581A · kind A · utility
Assignee
Inventor
Key dates
| Filing date | Apr 6, 1998 |
| Grant date | Dec 4, 2001 |
| Priority date | — |
| Expiry date | Apr 6, 2018 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F18/2411
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Solving a quadratic programming problem involved in training support vector machines by sweeping through a set of training examples, solving small sub-problems of the quadratic programming problem. Each of these sub-problems has an analytic solution, which is faster that the numerical quadratic programming solutions used in the prior art. In one embodiment, training examples with non-optimal Lagrange multipliers are adjusted, one at a time, until all are optimal (e.g. until all examples fulfill the Kuhn-Tucker conditions). In another embodiment, training examples with non-optimal Lagrange multipliers are paired and then adjusted, until all are optimal.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.