Tomaso Poggio

Tomaso Poggio is an Italian-born scientist known for foundational work at the intersection of computational neuroscience, artificial intelligence, and machine learning. He is a senior investigator and professor associated with the McGovern Institute for Brain Research and the Massachusetts Institute of Technology where he directs the Center for Biological and Computational Learning. His research connects theories from neuroscience, computer vision, and statistical learning theory to practical algorithms in computer science and cognitive science.

Early life and education

Poggio was born in Padua and educated in Italy with degrees from the University of Padua and the Scuola Normale Superiore di Pisa, institutions associated with figures like Enrico Fermi and Giuseppe Occhialini. He pursued early postdoctoral work at the European Molecular Biology Laboratory and collaborated with researchers from Istituto Nazionale di Fisica Nucleare and the Italian National Research Council. During his formative years he trained in environments tied to scientists such as Rita Levi-Montalcini and contemporaries in theoretical physics and biophysics.

Academic and research career

Poggio joined the faculty of the Massachusetts Institute of Technology and established the Center for Biological and Computational Learning and later became a senior scientist at the McGovern Institute for Brain Research. He has held visiting positions and collaborations with institutions including Stanford University, Harvard University, the European Molecular Biology Laboratory, and the Max Planck Society. His lab has hosted researchers from MIT Media Lab, Broad Institute, Harvard Medical School, and companies in the Silicon Valley ecosystem. He has served on advisory boards and committees for organizations such as the National Institutes of Health, the National Science Foundation, DARPA, and the European Research Council.

Contributions to computational neuroscience and AI

Poggio developed theoretical frameworks bridging neuroscience and machine learning including work on hierarchical models inspired by the visual cortex, studies of representation learning associated with Hubel and Wiesel-style receptive fields, and formal connections to statistical learning theory from researchers like Vladimir Vapnik and Alexey Chervonenkis. He proposed architectures and algorithms for object recognition and visual cortex modeling that influenced later systems such as convolutional neural network variants and contributed to the theoretical understanding behind deep learning made prominent by groups at Google DeepMind, OpenAI, and Facebook AI Research. His research on learning rules, regularization, and generalization relates to mathematical work by Geoffrey Hinton, Yann LeCun, Yoshua Bengio, and theorists studying the bias–variance tradeoff and kernel methods like Bernhard Schölkopf.

In computational neuroscience Poggio proposed mechanistic models for human vision and motor control tying experiments from labs of David Hubel, Torsten Wiesel, and Semir Zeki to computational theories. He advanced methods for linking neurophysiological data from techniques such as electrophysiology and functional magnetic resonance imaging used at centers like the Allen Institute for Brain Science to computational models. His group has produced algorithms used in collaborations with industry partners including IBM Research, Microsoft Research, and NVIDIA.

Awards and honors

Poggio's recognitions include election to academies and prizes from institutions such as the National Academy of Sciences, the American Academy of Arts and Sciences, the European Academy of Sciences, and receiving awards associated with societies like the Society for Neuroscience and the Association for Computing Machinery. He has been invited to give named lectures and has held fellowships tied to organizations including the Alexander von Humboldt Foundation and the Gordon and Betty Moore Foundation. His honors reflect contributions crossing communities represented by IEEE, SIAM, and JST-funded initiatives.

Selected publications and mentorship

Poggio authored influential papers and book chapters on topics spanning vision science, learning theory, and computational models published in venues connected to Nature, Science, Proceedings of the National Academy of Sciences, and conferences such as NeurIPS, ICML, and CVPR. His notable collaborators and coauthors include Vladimir N. Vapnik, Geoffrey Hinton, Yann LeCun, Kenji Doya, Christof Koch, and Shimon Ullman. He has supervised doctoral students and postdoctoral fellows who have become faculty and leaders at institutions including MIT, Stanford University, Harvard University, UC Berkeley, Caltech, ETH Zurich, and companies like Google, DeepMind, and OpenAI.

Category:Italian scientists Category:Computational neuroscientists Category:Massachusetts Institute of Technology faculty