Josh Tenenbaum

Papers

Total Citations

133

H-Index

About

Josh Tenenbaum is a leading figure in computational cognitive science, whose work bridges the gap between human intelligence and artificial intelligence. His core research focuses on understanding how the brain learns and infers the structure of the world, particularly through the lens of probabilistic models and Bayesian inference. Tenenbaum's major contributions include pioneering the "Bayesian brain" hypothesis, which posits that human cognition can be understood as a form of probabilistic reasoning over mental models. His influential work on "World model learning and inference" (81 citations) explores how the brain builds and uses internal models for high-level cognition and control, a key step toward general-purpose AI. In robotics, his "Residual Policy Learning" (52 citations) offers a practical method for improving imperfect robotic controllers through deep reinforcement learning, demonstrating how model-based and model-free approaches can be combined. With thousands of citations across his body of work, Tenenbaum's research has profoundly shaped our understanding of human learning and reasoning, and continues to inspire new approaches in AI and cognitive science.