Alberto Testolin

I am broadly interested in artificial intelligence, machine learning and cognitive neuroscience. In particular, I study how information is represented and processed in neural systems by means of computational simulations based on deep learning, recurrent neural networks and probabilistic graphical models.

In the cognitive neuroscience field, my research spans the domains of statistical learning, visual perception, orthographic processing, visuospatial attention, integration of bottom-up and top-down processing, predictive coding, numerical reasoning and mathematical learning.

In the engineering field, my research covers the theoretical and technological aspects of machine learning. In particular, I study the application of machine learning methods to signal processing, multidimensional data analysis and optimization of telecommunication systems. I am also interested in parallel computing architectures and advanced programming platforms for deep learning.

I also collaborate with physicists to better understand the underlying mechanisms that allow deep learning systems to be so effective. To this aim, we use methods from statistical mechanics and network theory to investigate the structural and functional properties of neural networks.

I am a member of the IEEE Task force on Deep Learning.

Education
Ph.D., Cognitive Science, University of Padova (2015)
Laurea (M.Sc.), Computer Science (Artificial Intelligence), University of Padova (2011)

Representative publications (see full list and PDF files here)

• Testolin, A., Kipnis, D., & Diamant, R. (2021). Detecting submerged objects using active acoustics and deep neural networks: A test case for pelagic fish. IEEE Transactions on Mobile Computing.
• Testolin A, Zou W, and McClelland, J (2020). Numerosity discrimination in deep neural networks: Initial competence, developmental refinement and experience statistics. Developmental Science.
• Testolin A, Piccolini M, and Suweis, S (2019). Deep learning systems as complex networks. Journal of Complex Networks.
• Testolin A, Stoianov I, and Zorzi M (2017). Letter perception emerges from unsupervised deep learning and recycling of natural image features. Nature Human Behaviour.
• Testolin, A., and Zorzi, M. (2016). Probabilistic models and generative neural networks: towards a unified framework for modeling normal and impaired neurocognitive functions. Frontiers in Computational Neuroscience.
• Testolin, A., Stoianov, I., Sperduti, A., and Zorzi, M. (2015). Learning orthographic structure with generative neural networks. Cognitive Science.