Tommaso Salvatori

Papers

Total Citations

H-Index

About

Tommaso Salvatori is a computational neuroscience and artificial intelligence researcher whose work sits at the fascinating intersection of brain-inspired computing and deep learning. His research centers on **predictive coding**, a biologically plausible framework that challenges the dominance of backpropagation — the algorithm that has long powered modern deep learning. Salvatori's most influential contributions explore whether predictive coding can serve as a viable, neurologically realistic alternative to backpropagation, addressing longstanding criticisms that standard training methods bear little resemblance to how biological brains actually learn. His 2022 paper, "Predictive Coding: Towards a Future of Deep Learning beyond Backpropagation?", has garnered 25 citations and positions predictive coding as a scalable, parallelizable framework better suited to both biological plausibility and next-generation hardware. His subsequent 2023 work on brain-inspired computational intelligence further develops these ideas, accumulating 12 citations and broadening the conversation around biologically grounded AI. Collectively, Salvatori's research challenges foundational assumptions in the field, offering a compelling vision for how future AI systems might learn in ways that more faithfully mirror the human brain — a direction of growing importance as the field seeks more efficient and interpretable learning algorithms.