About Joshua's Work
Joshua Tenenbaum is a cognitive scientist working at the intersection of computational cognitive science, developmental psychology, neuroscience, and artificial intelligence to investigate how the mind works. The human mind can acquire knowledge from very limited input and rapidly sort through noisy, conflicting information to reach judgments. Tenenbaum is one of the first to develop and apply probabilistic and statistical modeling to the study of human learning, reasoning, and perception, and to show how these models can explain a fundamental challenge of cognition: how our minds understand so much from so little, so quickly.
He and his group have demonstrated that the underlying cognitive and neural mechanisms that give rise to mental capabilities such as categorization, causal inference, and concept and language learning mirror Bayesian statistical approaches. Such approaches define the probability of a random event as determined by our prior beliefs and observed evidence rather than by the frequency (or pattern) by which it occurs. For example, Tenenbaum has shown that people’s estimates of the duration of an event, given just a single observed time point, closely correspond to the predictions of optimal statistical models, and his computational models of how children learn words from just one or a few examples have yielded insights into how children acquire their language skills. Tenenbaum supports his theoretical work with behavioral experiments with children and adults that demonstrate the effectiveness of his models in making quantitative predictions of human behavior. He has also used video game simulation engines and probabilistic programs to model how the mind constructs its intuitive physics and psychology—that is, how people develop an intuitive understanding of the ways physical objects and intentional agents interact in the world.
Tenenbaum has recently broadened his focus to the implications of his work for artificial intelligence and machine learning, which traditionally require much larger data sets for learning than humans typically employ. Tenenbaum is deepening our understanding of how the mind works and is making important contributions to bringing artificial intelligence closer to the capabilities of human cognition.
Joshua Tenenbaum received a BS (1993) from Yale University and a PhD (1999) from the Massachusetts Institute of Technology. He taught at Stanford University beginning in 1999 before returning to MIT in 2002. Tenenbaum currently serves as a professor in the Department of Brain and Cognitive Sciences at MIT, as a principal investigator in the Computer Science and Artificial Intelligence Lab (CSAIL), and as a research leader in MIT’s Center for Brains, Minds, and Machines. His articles have been published in Science, Trends in Cognitive Sciences, Psychological Review, PNAS, and Behavioral and Brain Sciences, among other journals.
Published on September 25, 2019