Marvin Faix

Centre National de la Recherche Scientifique, Laboratoire d'Informatique de Grenoble

Papers

Total Citations

H-Index

About

Marvin Faix is a researcher at the intersection of artificial intelligence, probabilistic programming, and cognitive science. His work explores how artificial systems can manage uncertainty through probabilistic models, drawing inspiration from natural cognition. Faix’s most cited paper (2017, 16 citations) investigates how probabilistic programming and stochastic arithmetic enable approximate computations with minimal resources, proposing these as plausible models for biological cognition. He extends this line of inquiry in a 2019 follow-up (8 citations), further developing automatic design methods for probabilistic systems. Faix also applies these theoretical foundations to robotics, as demonstrated in his work on stochastic Bayesian computation for autonomous robot sensorimotor systems (2015, 5 citations). His research bridges computational theory and practical implementation, offering insights into how machines can reason under uncertainty while mimicking cognitive efficiency. Faix’s contributions are particularly relevant for students and researchers interested in probabilistic AI, cognitive modeling, and resource-constrained robotics. His work underscores the potential of stochastic methods to create more adaptive, human-like artificial intelligence.