University of Bamberg
open search form
You are here

Information Systems and Applied Computer Sciences

Chair of Explainable Machine Learning

  1.  University of bamberg
  2. Faculties
  3. Information Systems and Applied Computer Sciences
  4. Subject groups
  5. Applied Computer Science
  6. Chair of Explainable Machine Learning
  7. Team
  8. Francesco Di Salvo, M.Sc.

Francesco Di Salvo

Tim Kipphan/Uni Bamberg

Teaching and Research Assistant

M.Sc., Doctoral Candidate

Adress:       An der Weberei 5, 96047 Bamberg
Room:             WE5/04.091

Phone:          +49-951-863 3107

Email:             francesco.di-salvo@uni-bamberg.de

Consultation hour:
during lecture times: Thursday 1 p.m. to 2 p.m. in WE5/04.091 ( notification via Email recommended)

outside lecture times: by appointment

Biography

 

Francesco Di Salvo is a PhD candidate at the Otto-Friedrich University of Bamberg, focused in the field of Explainable Machine Learning (xAI) since 2023. He holds a Bachelor of Science in Computer Engineering from the University of Palermo (Italy), and a Master of Science in Data Science and Engineering from the Polytechnic University of Turin (Italy).

His research interests lie in the intersection of model- and data-efficiency, explainability and robustness.

Before joining the xAI lab, between 2022 and 2023, Francesco worked as a research intern at NATO Center for Maritime Research and Experimentation (La Spezia, Italy), as part of the Young Scientist Internship Programme. He designed machine learning and deep learning algorithms to detect anomalies on underwater acoustic data.

Shortly before, he defended his Master's thesis, focused on Bayesian uncertainty on Contrast-Enhanced Breast CT images. The project was conducted in collaboration with the AXTI Lab of the Radboud University Medical Center (Nijmegen, Netherlands).

Furthermore, as a curricular internship, he spent four months as a Deep Learning Engineer Intern in AIKO - Infinite Ways of Autonomy, an Italian startup working on the autonomous space mission industry. He estimated the trajectory of a rover through monocular visual odometry methods, on planetary-like environments.

In his spare time, Francesco volunteered in academic associations, where he conducted monthly tutoring classes of calculus, linear algebra, and programming. He also mentored and coached incoming freshmen for their admission exams. Additionally, he served as a volunteer Machine Learning Engineer for Omdena, where he co-led a team of more than 20 volunteers. They developed machine learning algorithms for quantifying the impact of forest landscape restoration projects.

Outside of his professional activities, Francesco enjoys reading, cooking, and spending time outdoors playing volleyball, hiking, and doing any kind of water sports.

Profiles: Google scholar, ORCID, personal website, GitHub, LinkedIn, X

2023 - Today                                         PhD Candidate at the xAI Lab of the Otto-Friedrich University Bamberg, Germany
2022-2023Research Internship in anomaly detection for underwater acoustic data at NATO’s Center for Maritime Research and Experimentation (La Spezia, Italy)
2022

Research Internship in Bayesian uncertainty on contrast-enhanced breast CT scans at the Advanced X-ray Tomographic Imaging lab (Radboud University Medical Center, Nijmegen, Netherland

2021-2022

Curricular internship in Deep Learning for Monocular Visual Odometry in planetary-like environments at AIKO - Infinite Ways of Autonomy (Turin, Italy)

2020-2022Master of Science in Data Science and Engineering at Polytechnic University of Turin, Italy
2017-2020

Bachelor of Science in Computer Engineering at University of Palermo, Italy

 

Main Research Interests

  • Investigating the trade-off between model complexity and interpretability
  • Developing robust models toward corrupted, noisy and out-of-distribution data
  • Solving critical issues in Medical Image Analysis as lack of annotations, generalizability, model unlearning, interpretability and many more
  • Harnessing the strengths of foundation models for efficient utilization in downstream tasks

2024

  • Sebastian Doerrich, Tobias Archut, Francesco Di Salvo and Christian Ledig. "Integrating kNN with Foundation Models for Adaptable and Privacy-Aware Image Classification". 2024 IEEE International Symposium on Biomedical Imaging (ISBI). 2024 Paper Preprint Code

2023

  • Sebastian Doerrich, Francesco Di Salvo, and Christian Ledig. "unORANIC: Unsupervised Orthogonalization of Anatomy and Image-Characteristic Features". Machine Learning in Medical Imaging (MLMI). 2023. Paper Preprint Code

Master's Thesis

  • Francesco Di Salvo. "Deep learning for breast cancer diagnosis in contrast-enhanced breast CT". Master's thesis, Polytechnic University of Turin (2022, unpublished)

Thesis Supervision

Please check out our official bidding for thesis topics [Link to VC-Course] or contact me directly via email to request supervision of your Bachelor’s or Master’s Thesis.

WS24/25

  • xAI-Proj-M: Master Project Explainable Machine Learning
  • xAI-MML-M: General Exercises / Exam Preparation

SS24

  • xAI-Proj-M: Master Project Explainable Machine Learning
  • xAI-MML-M: Mathematics of Machine Learning - Exercises

WS23/24

  • xAI-DL-M: Deep Learning Exercises
  • xAI-Proj-M: Robust ML algorithms for real-world challenges

SS23

  • xAI-Sem-M: Masterseminar Explainable Machine Learning
  • xAI-MML-M: Mathematics of Machine Learning - Exercises

 

 

  • 2022: Selected as top candidate for the Young Scientist Internship Programme at NATO Centre for Maritime Research and Experimentation.
  • 2022: Full scholarship for completing the master thesis at the Radboud University Medical Center (Nijmegen, Netherlands).
  • 2020 - 2022: Two full scholarships based on academic track record (Polytechnic University of Turin).
  • 2017 - 2020: Three full scholarships based on academic track record (University of Palermo)