Jean-Pascal Pfister

University of Bern, University of Zurich

Papers

Total Citations

H-Index

About

Jean-Pascal Pfister is a leading researcher in computational neuroscience and statistical learning, with a focus on bridging theoretical frameworks and neural mechanisms. His primary research areas include online parameter estimation for stochastic processes, reinforcement learning in cortical networks, and the development of biologically plausible learning rules. Pfister’s major contribution lies in advancing the online maximum-likelihood estimation of partially observed diffusion processes, a critical tool for modeling dynamic systems where hidden states evolve continuously. His 2018 paper on this topic, with 20 citations, provides a recursive framework for real-time parameter updates, offering practical solutions for fields like neuroengineering and adaptive control. Additionally, his 2014 work on reinforcement learning in cortical networks, though less cited, explores how neural circuits implement reward-based learning, connecting theoretical models to biological reality. Pfister’s research is notable for its mathematical rigor and relevance to both machine learning and neuroscience, making his work essential for students and researchers interested in online learning algorithms, neural computation, and the intersection of statistics with biological systems. His contributions continue to influence adaptive filtering and neural network theory.