Brain & AI Lab


  • MatLab tools for building Mixture of Gaussian Scale Mixture (MGSM) models, and performing inference and learning. By Ruben Coen-Cagli, Peter Dayan, Odelia Schwartz.
    Code available on the CRCNS data sharing website: Code
    Associated paper: MGSM paper
    Cortical Surround Interactions and Perceptual Salience Via Natural Scene Statistics. R Coen-Cagli, P Dayan, and O Schwartz. PLoS Computational Biology, 8(3) 2013.
  • Python code sample for V2 textures and Deep Neural Networks: Code
    This is associated with the following ArXiv paper: Correspondence of Deep Neural Networks and the Brain for Visual Textures” Md Nasir Uddin Laskar, Luis G Sanchez Giraldo, and Odelia Schwartz, ArXiv Preprint 2018.
  • Matlab tutorial code for Spike-triggered Covariance: STCtutorial.tar.gz
    Based on Spike-triggered Neural Characterization. O Schwartz, J W Pillow, N C Rust, and E P Simoncelli. Journal of Vision, 2006. The code includes example simulations for model neurons and Gaussian inputs, with a choice of several model neurons (simple cell, complex cell, divisive normalization), and then code for the Spike Triggered Average and Covariance, and plotting of the eigenvalues and eigenvectors (estimated filters), and some simple plots of the nonlinearities. The best way to go through it is line by line, copying each time portions from the m file into Matlab.