Bruno Olshausen



  • Fellow
  • Learning in Machines & Brains


  • University of California Berkeley
Redwood Center for Theoretical Neuroscience and School of Optometry


  • United States


PhD (Computation and Neural Systems), California Institute of Technology
MS (Electrical Engineering), Stanford University
BS (Electrical Engineering), Stanford University


Neuroscientist Bruno Olshausen focuses on computational models of the visual cortex.

He is currently working on unsupervised learning principles that are capable of factoring time-varying images into independent representations of 3D form and motion. The hope is that such models can provide insight into the different forms of processing that occur in the ventral and dorsal streams of visual cortex.

Olshausen is co-founder of IQ Engines, a start-up that developed image search algorithms based on visual representations used in the brain, which was acquired by Yahoo! in 2013. He serves as advisor to Intel/Nervana, Vicarious and BayLabs.

Relevant Publications

Cheung B., E. Weiss, and B.A. Olshausen. "Emergence of foveal image sampling from learning to attend in visual scenes.” Paper presented at International Conference on Learning Representations (ICLR) conference, Toulon, France, 2017.

Lewicki, M.S. et al. "Scene analysis in the natural environment." Frontiers in Psychology 5 (2014).

Cadieu, C.F., and B.A. Olshausen. "Learning intermediate-level representations of form and motion from natural movies." Neural Computation 24, no. 4 (2012): 827–66.

Rozell, C.J. et al. "Sparse Coding via Thresholding and Local Competition in Neural Circuits." Neural Computation 20 (2008): 2526–2563.

Olshausen, B.A., and D.J. Field. "How Close Are We to Understanding V1?" Neural Computation 17 (2005): 1665–1699.

Rao, R.P.N. et al., eds. Probabilistic Models of the Brain: Perception and Neural Function. Cambridge, MA: MIT Press, 2002.