| Texto completo | |
| Autor(es): |
Alvarenga e Silva, Lucas Fernando
;
Sebe, Nicu
;
Almeida, Jurandy
Número total de Autores: 3
|
| Tipo de documento: | Artigo Científico |
| Fonte: | 2023 36TH CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES, SIBGRAPI 2023; v. N/A, p. 6-pg., 2023-01-01. |
| Resumo | |
Convolutional Neural Networks (CNNs) have brought revolutionary advances to many research areas due to their capacity of learning from raw data. However, when those methods are applied to non-controllable environments, many different factors can degrade the model's expected performance, such as unlabeled datasets with different levels of domain shift and category shift. Particularly, when both issues occur at the same time, we tackle this challenging setup as Open Set Domain Adaptation (OSDA) problem. In general, existing OSDA approaches focus their efforts only on aligning known classes or, if they already extract possible negative instances, use them as a new category learned with supervision during the course of training. We propose a novel way to improve OSDA approaches by extracting a high-confidence set of unknown instances and using it as a hard constraint to tighten the classification boundaries of OSDA methods. Especially, we adopt a new loss constraint evaluated in three different means, (1) directly with the pristine negative instances; (2) with randomly transformed negatives using data augmentation techniques; and (3) with synthetically generated negatives containing adversarial features. We assessed all approaches in an extensive set of experiments based on OVANet, where we could observe consistent improvements for two public benchmarks, the Office-31 and Office-Home datasets, yielding absolute gains of up to 1.3% for both Accuracy and H-Score on Office-31 and 5.8% for Accuracy and 4.7% for H-Score on Office-Home. (AU) | |
| Processo FAPESP: | 17/25908-6 - Aprendizado fracamente supervisionado para análise de vídeos no domínio comprimido em tarefas de recuperação e classificação para alertas visuais |
| Beneficiário: | João Paulo Papa |
| Modalidade de apoio: | Auxílio à Pesquisa - Parceria para Inovação Tecnológica - PITE |
| Processo FAPESP: | 23/03328-9 - Esmiuçando o Aprendizado de Máquina: Aceleradores Eficientes para Redes Profundas e sua Aplicabilidade em Computação Cientifica |
| Beneficiário: | Lucas Fernando Alvarenga e Silva |
| Modalidade de apoio: | Bolsas no Brasil - Doutorado |
| Processo FAPESP: | 20/08770-3 - Métodos de reconhecimento multimídia com uso de redes profundas para conjuntos abertos |
| Beneficiário: | Lucas Fernando Alvarenga e Silva |
| Modalidade de apoio: | Bolsas no Brasil - Mestrado |
| Processo FAPESP: | 21/13348-1 - Investigação de métodos de adaptação de domínio de conjunto aberto para problemas de visão computacional |
| Beneficiário: | Lucas Fernando Alvarenga e Silva |
| Modalidade de apoio: | Bolsas no Exterior - Estágio de Pesquisa - Mestrado |