Publications

Publications by Carlos Manuel Soares

2018

Towards Reproducible Empirical Research in Meta-Learning

Authors
Rivolli, A; Garcia, LPF; Soares, C; Vanschoren, J; de Carvalho, ACPLF;

Publication
CoRR

Abstract

2018

Algorithm Selection for Collaborative Filtering: the influence of graph metafeatures and multicriteria metatargets

Authors
Cunha, T; Soares, C; de Carvalho, ACPLF;

Publication
CoRR

Abstract

2018

Building robust prediction models for defective sensor data using Artificial Neural Networks

Authors
Shekar, AK; de Sá, CR; Ferreira, H; Soares, C;

Publication
CoRR

Abstract

2018

Smart energy management as a means towards improved energy efficiency

Authors
Lindert, Dt; de Sá, CR; Soares, C; Knobbe, AJ;

Publication
CoRR

Abstract

2019

Preference rules for label ranking: Mining patterns in multi-target relations

Authors
de Sá, CR; Azevedo, PJ; Soares, C; Jorge, AM; Knobbe, AJ;

Publication
CoRR

Abstract

2023

GASTeN: Generative Adversarial Stress Test Networks

Authors
Cunha, L; Soares, C; Restivo, A; Teixeira, LF;

Publication
ADVANCES IN INTELLIGENT DATA ANALYSIS XXI, IDA 2023

Abstract
Concerns with the interpretability of ML models are growing as the technology is used in increasingly sensitive domains (e.g., health and public administration). Synthetic data can be used to understand models better, for instance, if the examples are generated close to the frontier between classes. However, data augmentation techniques, such as Generative Adversarial Networks (GAN), have been mostly used to generate training data that leads to better models. We propose a variation of GANs that, given a model, generates realistic data that is classified with low confidence by a given classifier. The generated examples can be used in order to gain insights on the frontier between classes. We empirically evaluate our approach on two well-known image classification benchmark datasets, MNIST and Fashion MNIST. Results show that the approach is able to generate images that are closer to the frontier when compared to the original ones, but still realistic. Manual inspection confirms that some of those images are confusing even for humans.

CloseRead Abstract