Riccardo Zecchina

Professor, Theoretical Physics
Bocconi University.

Still collaborating with:
Department of Applied Science and Technology
Institute of Condensed Matter Physics and Complex Systems
Politecnico di Torino
Corso Duca degli Abruzzi 24, 10129 Torino (Italy)

Head of Unit at the Human Genetics Foundation
Statistical Inference and Computational Biology.

Current research interests:

Statistical physics, machine learning, computational neuroscience and computational biology, inverse dynamical problems, distributed algorithms for optimisation, constraint satisfaction problems and statistical inference, information theory, graphical games, …

An almost complete list of preprints from the ArXiv. Links to papers and codes on the sidebar.

Recent papers and some preprints

  • Inverse statistical problems: from the inverse Ising problem to data science, HC Nguyen, R Zecchina, J Berg, arXiv preprint arXiv:1702.01522
  • Unreasonable effectiveness of learning neural networks: From accessible states and robust ensembles to basic algorithmic schemes, C Baldassi, C Borgs, JT Chayes, A Ingrosso, C Lucibello, L Saglietti, R. Zecchina Proceedings of the National Academy of Sciences 113 (48), E7655-E76621
  • Learning may need only a few bits of synaptic precision, C Baldassi, F Gerace, C Lucibello, L Saglietti, R Zecchina, Physical Review E 93 (5), 052313
  • Subdominant dense clusters allow for simple learning and high computational performance in neural networks with discrete synapses, C. Baldassi, A. Ingrosso, C. Lucibello, L. Saglietti, R. Zecchina, Phys. Rev. Lett. 115, 128101 (2015)
  • A three-threshold learning rule approaches the maximal capacity of recurrent neural networks, A. Alemi, C. Baldassi, N. Brunel, R. Zecchina, PLoS Comput Biol 11, e1004439 (2015)
  • Quantitative study of crossregulation, noise and synchronization between microRNA targets in single cells, C. Bosia, F. Sgrò, L. Conti, C. Baldassi, F. Cavallo, F. Di Cunto, E. Turco, A. Pagnani, R. Zecchina, under review
  • Local entropy as a measure for sampling solutions in constrained satisfaction problems, C. Baldassi, A. Ingrosso, C. Lucibello, L. Saglietti, R. Zecchina, Journal of Statistical Mechanics: Theory and Experiment 2016 (2), 023301
  • Entropy-SGD: Biasing Gradient Descent Into Wide Valleys, Pratik Chaudhari, Anna Choromanska, Stefano Soatto, Yann LeCun, Carlo Baldassi, Christian Borgs, Jennifer Chayes, Levent Sagun, Riccardo Zecchinarx, arXiv:1611.01838

Multidisciplinary projects in our labs

Machine Learning
  • Nonequilibrium statistical physics approach to learning and deep learning
  • Complexity of learning in neural circuits with constrained “material” synapses
  • Modelling in Neuromorphic engineering: message-passing in situ hardware learning for large scale data analysis

Statistical Inference and Computational Biology Unit, Human Genetics Foundation (HuGeF)
  • Modelling post-transcriptional regulation: theory and experiments on miRNA/mRNAs interaction
  • Protein contacts inference from evolutionary sequence data
  • Inverse problems arising in Single Cell Genomics