Nearest neighbors distance ratio open-set classifier

Autor(es)

Mendes Junior, Pedro R. ; de Souza, Roberto M. ; Werneck, Rafael de O. ; Stein, Bernardo V. ; Pazinato, Daniel V. ; de Almeida, Waldir R. ; Penatti, Otavio A. B. ; Torres, Ricardo da S. ; Rocha, Anderson

• Tipo de documento: Artigo Científico
• Fonte: MACHINE LEARNING; v. 106, n. 3, p. 359-386, MAR 2017.
• Citações Web of Science: 8

Resumo

In this paper, we propose a novel multiclass classifier for the open-set recognition scenario. This scenario is the one in which there are no a priori training samples for some classes that might appear during testing. Usually, many applications are inherently open set. Consequently, successful closed-set solutions in the literature are not always suitable for real-world recognition problems. The proposed open-set classifier extends upon the Nearest-Neighbor (NN) classifier. Nearest neighbors are simple, parameter independent, multiclass, and widely used for closed-set problems. The proposed Open-Set NN (OSNN) method incorporates the ability of recognizing samples belonging to classes that are unknown at training time, being suitable for open-set recognition. In addition, we explore evaluation measures for open-set problems, properly measuring the resilience of methods to unknown classes during testing. For validation, we consider large freely-available benchmarks with different open-set recognition regimes and demonstrate that the proposed OSNN significantly outperforms their counterparts in the literature. (AU)

• Processo FAPESP: 10/05647-4 - Computação forense e criminalística de documentos: coleta, organização, classificação e análise de evidências
• Beneficiário: Anderson de Rezende Rocha
• Modalidade de apoio: Auxílio à Pesquisa - Jovens Pesquisadores
• Processo FAPESP: 13/50155-0 - Combining new technologies to monitor phenology from leaves to ecosystems
• Beneficiário: Leonor Patricia Cerdeira Morellato
• Modalidade de apoio: Auxílio à Pesquisa - Programa de Pesquisa sobre Mudanças Climáticas Globais - PITE
• Processo FAPESP: 15/19222-9 - DejaVu: análise forense de mídias sociais para interpretação de eventos criminais
• Beneficiário: Anderson de Rezende Rocha
• Modalidade de apoio: Bolsas no Exterior - Pesquisa
• Processo FAPESP: 13/50169-1 - Towards an understanding of tipping points within tropical South American biomes
• Beneficiário: Ricardo da Silva Torres
• Modalidade de apoio: Auxílio à Pesquisa - Parceria para Inovação Tecnológica - PITE