Carlo Baldassi
Carlo Baldassi studies the mathematical structure of neural-network learning. His work connects statistical mechanics to optimization and computational neuroscience. Baldassi studied theoretical physics in Trieste and earned a doctorate in computational neuroscience in Turin. He later focused on machine learning and neural networks. He coauthored research on dense clusters of solutions in neural networks with discrete synapses. The Physical Review Letters paper studies how such solution structure relates to learning performance. He also coauthored research on shaping neural-network learning landscapes around wide flat minima. He studies loss landscapes analytically and numerically.
- Now at
- Bocconi University
- Works
- 74
- Citations
- 1,861
- h-index
- 22
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2015
Learning with discrete connections
He coauthored research on dense clusters of solutions in neural networks with discrete synapses. The Physical Review Letters paper studies how such solution structure relates to learning performance.
-
2020
The geometry of learning
He also coauthored research on shaping neural-network learning landscapes around wide flat minima. He studies loss landscapes analytically and numerically.
-
2023–2024 academic year
AI degree leadership
Baldassi led Bocconi’s bachelor’s degree in Mathematical and Computing Sciences for Artificial Intelligence from the 2023–2024 academic year.
-
As of October 2026
Physics and neural computation
Baldassi studied theoretical physics in Trieste and earned a doctorate in computational neuroscience in Turin. He later focused on machine learning and neural networks.
Sources and credits
Photo: ellis.eu; Carlo Baldassi profile photograph; creator not stated. License: Not stated by source. Source: ellis.eu. Cropped to a square and resized. Ask us to remove or replace this photo.