Dmitri Chklovskii

Research Associate Professor, Neuroscience & Physiology
Neuroscience Institute

NYU Neuroscience Institute
Alexandria Center for Life
450 East 29th St.
New York, NY 10016
Fax: or ;
Lab Website: website

Research Summary:

Neural computation and connectomics

The human brain makes sense of the world and comes up with new ideas with an efficiency, speed and accuracy unattainable by modern computers. Our goal is to understand how the brain analyzes large and complex datasets streamed by sensory organs in order to aid efforts at treating mental illness and building artificial neural systems. We analyze experimental data, assembling connectomes from high-throughput electron microscopy and determining neuronal dynamics from calcium imaging and multi-electrode recordings. Empowered by experimental observations, we are developing a novel algorithmic theory of neural computation.

Selected Publications:

  • C. Pehlevan, DB Chklovskii. 2015. A Normative theory of adaptive dimensionality reduction in neural networks. Neural Information Processing Systems (NIPS)
  • C. Pehlevan, T. Hu, DB Chklovskii. 2015 A Hebbain/Anti-Hebbian neural network for linear subspace learning: A derivation from multidimensional scaling of streaming data. Neural Computation 03/2015; 27(7).
  • Plaza, S, Scheffer, L, Chklovskii DB, (2014) Towards large-scale connectome reconstructions. Current Opinion in Neurobiology 25: 201-210.
  • Takemura SY, Bharioke A, Lu Z, Nern A, Vitaladevuni S, Rivlin P, Katz W, Olbris D, Plaza SM, Winston P, Zhao T, Horne JA, Fetter R, Takemura SK, Blazek K, Chang LA, Ogundeyi O, Saunders M, Shapiro V, Sigmund C, Rubin GM, Scheffer LK, Meinertzhagen IA, Chklovskii DB. (2013) A visual motion detection circuit suggested by Drosophila connectomics, Nature 500:175-181.
  • S. Druckmann & D.B. Chklovskii. 2012. Neuronal Circuits Underlying Persistent Representations Despite Time Varying Activity. Current Biology 22(4): 2095–2103.
  • T. Hu, A. Genkin, DB Chklovskii. 2012. A network of spiking neurons for computing sparse representations in an energy efficient way. Neural Computation, 24(11) 2852-2872.

Research Image: