IARPA-MICrONS: "Reverse Engineering Neocortical Intelligence"

15/07/2017 to 14/07/2019

Given the expertise of the Bethge group - in computational neuroscience, machine learning, statistics, and the analyses of complex neural data including methods of pre-processing calcium data (e. g. spike inference) - the team will work on a number of problems/questions regarding the goals and deliverables of Technical Area 1 in coordination with Drs. Tolias (PI), Pitkow (Co-PI), Reimer (Project Manager) and Sinz (ML coordinator).
1) Use system identification and other statistical methods to analyze functional imaging data.
2) Help with analysis of combined functional and electron microscopy data.
3) Work on adversarial attacks to engineer robust visual architectures inspired by the brain.
4) Work on neuroscience-inspired unsupervised learning algorithms.
5) Help with pipeline analysis methods using machine learning.