Brenden Lake

Assistant Professor
Department of Psychology and Center for Data Science (joint)
CILVR Lab (research group)
New York University

Contact

Email: brenden@nyu.edu
Phone: (212)998-3059
Primary office (Data Science): 60 5th Ave., Room 610, New York, NY 10011
Secondary office (Psychology): 6 Washington Place, Room 873B, New York, NY 10003
Curriculum Vitae (as of June 2017)

News

I am currently recruiting graduate students and a postdoctoral fellow. Interested graduate candidates can apply to either the Ph.D. Program in Cognition and Perception (deadline is Dec. 1, 2017) or the Ph.D. program in Data Science (deadline is Jan. 4, 2018). Please email me for more information.

Research

I build computational models of everyday cognitive abilities, focusing on problems that are easier for people than they are for machines. The human mind is the best known solution to a diverse array of difficult computational problems: learning new concepts, learning new tasks, understanding scenes, learning language, asking questions, forming explanations, amongst many others. Machines also struggle to simulate other facets of human intelligence, including creativity, curiosity, self-assessment, and commonsense reasoning.

In this broad space of computational challenges, my work has addressed a range of questions: How do people learn a new concept from just one or a few examples? How do people act creatively when designing new concepts? How do people learn qualitatively different forms of structure? How do people ask questions when searching for information?

By studying these distinctively human endeavors, there is potential to advance both cognitive science and machine learning. In cognitive science, building a computational model is a test of understanding; if people outperform all existing algorithms on certain types of problems, we have more to understand about how people solve them. In machine learning, these cognitive abilities are both important open problems as well as opportunities to reverse engineer the human solutions. By studying the human solutions to difficult computational problems, I aim to better understand humans and to build machines that learn in more powerful and more human-like ways.

Representative talk

Video of 2016 Stanford EE Computer Systems Colloquium

Preprints

Lake, B. M., Lawrence, N. D., and Tenenbaum, J. B. (2016). The emergence of organizing structure in conceptual representation. Preprint available on arXiv:1611.09384. [Code]

Publications

Lake, B. M., Ullman, T. D., Tenenbaum, J. B., and Gershman, S. J. (in press). Building machines that learn and think like people. Behavioral and Brain Sciences. Target Article.

Rothe, A., Lake, B. M., and Gureckis, T. M. (2016). Asking and evaluating natural language questions. In Proceedings of the 38th Annual Conference of the Cognitive Science Society.

Cohen, A. and Lake, B. M. (2016). Searching large hypothesis spaces by asking questions. In Proceedings of the 38th Annual Conference of the Cognitive Science Society.

Lake, B. M., Salakhutdinov, R., and Tenenbaum, J. B. (2015). Human-level concept learning through probabilistic program induction. Science, 350(6266), 1332-1338. [Supporting Info.] [visual Turing tests] [Omniglot data set] [Bayesian Program Learning code]

Monfort, M., Lake, B. M., Ziebart, B. D., Lucey, P., and Tenenbaum, J. B. (2015). Softstar: Heuristic-Guided Probabilistic Inference. Advances in Neural Information Processing Systems 28. [Supporting Info.]

Lake, B. M., Zaremba, W., Fergus, R. and Gureckis, T. M. (2015). Deep Neural Networks Predict Category Typicality Ratings for Images. In Proceedings of the 37th Annual Conference of the Cognitive Science Society. [Data]

Lake, B. M. (2014). Towards more human-like concept learning in machines: Compositionality, causality, and learning-to-learn. Ph.D. thesis, MIT.

Lake, B. M., Lee, C.-y., Glass, J. R., and Tenenbaum, J. B. (2014). One-shot learning of generative speech concepts. In Proceedings of the 36th Annual Conference of the Cognitive Science Society. [Supporting Info.]

Lake, B. M., Salakhutdinov, R., and Tenenbaum, J. B. (2013). One-Shot Learning by Inverting a Compositional Causal Process. Advances in Neural Information Processing Systems 26. [Supporting Info.]

Lake, B. M., Salakhutdinov, R., and Tenenbaum, J. B. (2012). Concept learning as motor program induction: A large-scale empirical study. In Proceedings of the 34th Annual Conference of the Cognitive Science Society. [Supporting Info.]

Lake, B. M., Salakhutdinov, R., Gross, J., and Tenenbaum, J. B. (2011). One shot learning of simple visual concepts. In Proceedings of the 33rd Annual Conference of the Cognitive Science Society. [Videos]

Lake, B. M. and McClelland, J. L. (2011). Estimating the strength of unlabeled information during semi-supervised learning. In Proceedings of the 33rd Annual Conference of the Cognitive Science Society.

Lake, B. M. and Tenenbaum, J. B. (2010). Discovering Structure by Learning Sparse Graphs. In Proceedings of the 32nd Annual Conference of the Cognitive Science Society.

Lake, B. M. (2009). Unsupervised and semi-supervised perceptual category learning. Master's thesis, Stanford University.

Lake, B. M., Vallabha, G. K., and McClelland, J. L. (2009). Modeling unsupervised perceptual category learning. IEEE Transactions on Autonomous Mental Development, 1(1), 35-43.

Lake, B. M., Vallabha, G. K., and McClelland, J. L. (2008). Modeling unsupervised perceptual category learning. In Proceedings of the 7th International Conference on Development and Learning. Best paper award. Expanded version directly above.

Lake, B. M. and Cottrell, G.W. (2005). Age of acquisition in facial identification: A connectionist approach. In Proceedings of the 27th Annual Conference of the Cognitive Science Society.


Other

Cognitive ToyBox