My morning routine is quite stereotyped: the first step is always to turn on my favorite coffee machine. I see this machine every day, effortlessly recognizing it as “my coffee machine”. Indeed, it looks the same to me each morning. While we often take this reliability of sensory perception for granted, neurophysiology experiments indicate that, beneath the veneer of robustness, the activities of our sensory neurons are highly variable. For example, if one repeatedly presents the same stimulus to an animal and records the activities of their sensory neurons, those neural responses typically vary significantly from trial-to-trial. These observations suggest the following questions: what adaptations allow our brains to function robustly despite being made of seemingly unreliable components?; and in what ways might internal variability enhance the brain’s function?
In my laboratory, we create mathematical and computational models of systems that can reliably encode sensory information, or compute relevant features of the environment, using components that show biologically-realistic levels of variability. These models allow us to identify key features that give rise to robust system function, leading to predictions for neurophysiology experiments. By collaborating with experimenters to test these predictions, my research program identifies the mechanisms through which the brain mitigates the potentially negative impacts of internal variability, and exploits its benefits to better make inferences about the outside world.
N.A. Cayco-Gajic, J. Zylberberg, and E. Shea-Brown (2015). Triplet correlations among similarly tuned cells impact population coding. Frontiers in Computational Neuroscience 9: 57.
Y. Hu, J. Zylberberg, and E. Shea-Brown (2014). The sign rule and beyond: Boundary effects, ﬂexibility, and noise correlations in neural population codes. PLoS Computational Biology 10: e1003469.
J. Zylberberg and M.R. DeWeese (2013). Sparse coding models can exhibit decreasing sparseness while learning sparse codes for natural images. PLoS Computational Biology 9: e1003182.
P. King, J. Zylberberg, and M.R. DeWeese (2013). Inhibitory interneurons decorrelate excitatory cells to drive sparse code formation in a spiking model of V1. Journal of Neuroscience 33: 5475-5485.
J. Zylberberg, J. Murphy, and M.R. DeWeese (2011). A Sparse Coding Model with Synaptically Local Plasticity and Spiking Neurons Can Account for the Diverse Shapes of V1 Simple Cell Receptive Fields. PLoS Computational Biology 7: e1002250.
J. Zylberberg, D. Pfau, and M.R. DeWeese (2012). Dead leaves and the dirty ground: low-level image statistics in transmissive and occlusive imaging environments. Physical Review E 86: 066112.
G. Zhao, L. Pogosian, A. Silvestri, and J. Zylberberg (2009). Cosmological Tests of General Relativity with Future Tomographic Surveys. Physical Review Letters 103: 241301.
J. Zylberberg et al. (2007). Charge-state distributions after radiative capture of helium nuclei by a carbon beam. Nuclear Instruments and Methods in Physics Research B 254: 17-24.
J. Zylberberg, A.A. Belik, E. Takayama-Muromachi, and Z.-G. Ye (2007). Bismuth Aluminate: A New High-TC Lead-Free Piezo-/ferroelectric. Chemistry of Materials 19: 6385-6390.
J. Zylberberg and Z.-G. Ye (2006). Improved dielectric properties of bismuth-doped LaAlO3. Journal of Applied Physics 100: 086102.
1. Howard Hughes Medical Institute International Student Research Fellowship (2011).
2. “Outstanding Graduate Student Instructor" award from UC Berkeley (2011).
3. Fulbright Science and Technology Ph.D. Fellowship (2008).
4. Natural Sciences and Engineering Research Council of Canada Julie Payette Research Scholarship (2008).