Detalhes
Nome
Bruno Fonseca CoelhoCargo
Assistente de InvestigaçãoDesde
15 maio 2023
Nacionalidade
BrasilCentro
Telecomunicações e MultimédiaContactos
+351222094000
bruno.f.coelho@inesctec.pt
2025
Autores
Oliveira Coelho, BF; Cardoso, JS;
Publicação
Neurocomputing
Abstract
In order to facilitate the adoption of deep learning in areas where decisions are of critical importance, understanding the model's internal workings is paramount. Nevertheless, since most models are considered black boxes, this task is usually not trivial, especially when the user does not have access to the network's intermediate outputs. In this paper, we propose IBISA, a model-agnostic attribution method that reaches state-of-the-art performance by optimizing sampling masks using the Information Bottleneck Principle. Our method improves on the previously known RISE and IBA techniques by placing the bottleneck right after the image input without complex formulations to estimate the mutual information. The method also requires only twenty forward passes and ten backward passes through the network, which is significantly faster than RISE, which needs at least 4000 forward passes. We evaluated IBISA using a VGG-16 and a ResNET-50 model, showing that our method produces explanations comparable or superior to IBA, RISE, and Grad-CAM but much more efficiently. © 2025 The Authors
2025
Autores
Massaranduba, ABR; Coelho, BFO; Santos Souza, CAd; Viana, GG; Brys, I; Ramos, RP;
Publicação
Current Psychology
Abstract
2024
Autores
Coelho, BFO; Nunes, SLP; de França, CA; Costa, DdS; do Carmo, RF; Prates, RM; Filho, EFS; Ramos, RP;
Publicação
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy
Abstract
2024
Autores
Souza, C; Viana, G; Coelho, B; Massaranduba, AB; Ramos, R;
Publicação
Anais do XVI Congresso Brasileiro de Inteligência Computacional
Abstract
2023
Autores
Massaranduba, AB; Coelho, B; Machado, E; Silva, E; Pinto, A;
Publicação
IEEE Latin America Transactions
Abstract
The access to the final selection minute is only available to applicants.
Please check the confirmation e-mail of your application to obtain the access code.