Fabio Augusto Faria

Title: Learning More for Less: Deep Metric Learning Approaches for Incomplete Supervision.

Abstract: Deep learning architectures (DL) have achieved promising results in different areas (e.g., medicine, agriculture, and security). However, using those powerful techniques in many real applications becomes challenging due to the large labeled collections required during training. Several works have pursued solutions to overcome it by proposing strategies that can learn more for less, e.g., supervised (transfer and few-shot learning), weakly (inexact, inaccurate, and incomplete supervision) and semi-supervised (label propagation and pseudo-labeling) learning approaches. As these approaches do not usually address memorization and sensitivity to adversarial examples of the DL, this paper presents new deep metric learning approaches combined with Mixup technique for incomplete-supervision scenarios. We show that some state-of-the-art approaches in deep metric learning might not work well in such scenarios. Moreover, the proposed approaches outperform most of them in different and well-known datasets. Finally, we employ our approaches for complete scenarios (image classification tasks), achieving excellent classification results compared to the original version approaches in the literature.
*Presentation based on the paper accepted for publication at ICIP 2022 - https://arxiv.org/pdf/2204.13572.pdf

Bio: Fabio A. Faria received a B.Sc. in Computer Science from the State University of Sao Paulo, Brazil, in 2007, his Master and Ph.D. degrees in Computer Science at the University of Campinas (Unicamp) in 2010 and 2014, respectively. From April/2012 to April/2013 he was a visiting scholar at the University of South Florida (USF) under supervision of Prof. Dr. Sudeep Sarkar. From July/2019 to September/2020, he was a visiting researcher at the Australian Institute of Machine Learning (AIML) from the University of Adelaide in Australia under supervision of Prof. Dr. Gustavo Carneiro. Since 2015, he is Professor at the Institute of Science and Technology from the Universidade Federal de São Paulo (UNIFESP). He has been developing multidisciplinary research projects involving Machine Learning, Information Fusion, Image Processing, and Computer Vision on the eScience domain.

Webpage - https://www.ict.unifesp.br/ffaria

Moderation: Prof. Jacinto Nascimento, Scientific Area of Graphics and Interaction (IG), Department of Computer Science and Engineering (DEI), Instituto Superior Técnico (IST)