Fabrizio Pastore

Papers

Total Citations

H-Index

About

Fabrizio Pastore is a leading researcher in software engineering, with a focus on automated testing, debugging, and the reliability of AI-based systems. His work bridges the gap between classical software testing and modern machine learning, particularly in safety-critical domains like autonomous vehicles and robotics. A major contribution is his pioneering work on search-based testing for Deep Neural Networks (DNNs), where he developed techniques to generate realistic, failure-inducing scenarios using GAN-enhanced simulations. This approach, detailed in his highly cited 2025 paper, enables automated retraining to improve DNN robustness without manual labeling. Pastore’s broader impact is reflected in over 2,000 citations, with key contributions to spectrum-based fault localization and metamorphic testing. He has also advanced the field through his work on test case generation for cyber-physical systems and continuous integration. His research is widely recognized for its practical relevance, earning him multiple best paper awards and a strong presence in top venues like ICSE and FSE. For students and researchers, Pastore’s work offers a compelling blueprint for ensuring the dependability of the next generation of intelligent, autonomous software.