Classification system trainer employing maximum margin back-propagation with probabilistic outputs
US6728690B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Nov 23, 1999 |
| Grant date | Apr 27, 2004 |
| Priority date | — |
| Expiry date | Nov 23, 2019 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N3/09
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A training system for a classifier utilizes both a back-propagation system to iteratively modify parameters of functions which provide raw output indications of desired categories, wherein the parameters are modified based on a weighted decay, and a probability determining system with further parameters that are determined during iterative training. A margin error metric may be combined with weight decay, and a sigmoid is used to calibrate the raw outputs to probability percentages for each category. A method of training such a system involves gathering a training set of inputs and desired corresponding outputs. Classifier parameters are then initialized and an error margin is calculated with respect to the classifier parameters. A weight decay is then used to adjust the parameters. After a selected number of times through the training set, the parameters are deemed in final form, and an optimization routine is used to derive a set of probability transducer parameters for use in calculating the probable classification for each input.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.