Adequacy Metric

Adequacy metrics assess the thoroughness and effectiveness of testing procedures, particularly for complex systems like deep learning models. Current research focuses on developing and comparing these metrics across various applications, including software testing, adversarial robustness in image classification, and fairness evaluations in image recognition. This work aims to improve the reliability and efficiency of testing by identifying which metrics best reveal weaknesses and guide improvements in model performance and ethical considerations. The ultimate goal is to enhance the trustworthiness and responsible deployment of these increasingly prevalent technologies.

Papers

October 11, 2023

Assessing Evaluation Metrics for Neural Test Oracle Generation
Jiho Shin, Hadi Hemmati, Moshi Wei, Song Wang
Evaluation Metric Test Oracle Adequacy Metric

July 8, 2022

Guiding the retraining of convolutional neural networks against adversarial inputs
Francisco Durán López, Silverio Martínez-Fernández, Michael Felderer, Xavier Franch
Convolutional Neural Network Deep Learning Model Adversarial Input Retraining Method Adequacy Metric

November 17, 2021

Fairness Testing of Deep Image Classification with Adequacy Metrics
Peixin Zhang, Jingyi Wang, Jun Sun, Xinyu Wang
Image Classification Fairness Perception FAIrness Testing Fairness Testing Adequacy Metric

Adequacy Metric

Papers

Assessing Evaluation Metrics for Neural Test Oracle Generation

Guiding the retraining of convolutional neural networks against adversarial inputs

Fairness Testing of Deep Image Classification with Adequacy Metrics