Understanding the human brain is one of the greatest and most challenging scientific frontiers of our time. CCN’s mission is to develop models, principles and conceptual frameworks that deepen our knowledge of brain function — both in health and in disease.
Featured News
Center for Computational Neuroscience
Biological Brains Inspire a New Building Block for Artificial Neural Networks
Read More
While artificial intelligence systems have advanced tremendously in recent years, they still lag behind the performance of real brains in reliability and efficiency. A new type of computational unit developed at the Flatiron Institute could help close that gap.
CCN takes a “systems" neuroscience approach, building models that are motivated by fundamental principles, that are constrained by properties of neural circuits and responses, and that provide insights into perception, cognition and behavior. This cross-disciplinary approach not only leads to the design of new model-driven scientific experiments, but also encapsulates current functional descriptions of the brain that can spur the development of new engineered computational systems, especially in the realm of machine learning. CCN currently has research groups in Computational Vision and Neural Circuits and Algorithms, and will launch research groups in NeuroAI and Geometry and Statistical Analysis of Neural Data in January 2022.
Research
Collaborative Work
Machine Learning at the Flatiron Institute
Machine learning has become an indispensable tool for computational science, and it is an active and growing area of research throughout the Flatiron Institute.
- CCM
- CCN
- CCA
- CCB
- CCQ
Upcoming Events
-
03 Tue
-
01 Wed -
Colloquium 4:00 - 8:30 p.m.
Joint Theory-Experiment Colloquia
-
Colloquium 4:00 - 8:30 p.m.
Joint Theory-Experiment Colloquia
-
Colloquium 4:00 - 8:30 p.m.
-
19 Tue -
Workshop
Control Theory & Neuroscience
-
Workshop
Publications
Generalized Compressed Sensing for Image Reconstruction with Diffusion Probabilistic Models
We examine the problem of selecting a small set of linear measurements for reconstructing high-dimensional signals. Well-established methods for optimizing…
Transactions on Machine Learning ResearchA Network of Biologically Inspired Rectified Spectral Units (ReSUs) Learns Hierarchical Features Without Error Backpropagation
We introduce a biologically inspired, multilayer neural architecture composed of Rectified Spectral Units (ReSUs). Each ReSU projects a recent window…
arXiv:2512.23146Disentangled representations via score-based variational autoencoders
Benjamin S. H. Lyo, C. Savin, E. P. Simoncelli
We present the Score-based Autoencoder for Multiscale Inference (SAMI), a method for unsupervised representation learning that combines the theoretical frameworks…
arXiv:2512.17127Leadership
Software
CaImAn Python
Computational toolbox for large scale Calcium Imaging Analysis, including movie handling, motion correction, source extraction, spike deconvolution and result visualization.
NeMoS
A statistical modeling framework for systems neuroscience. NeMos specializes in GPU-accelerated optimizations.
plenoptic
`plenoptic` is a python library for model-based stimulus synthesis.
PYthon Neural Analysis Package (Pynapple)
Pynapple is a light-weight python library for neurophysiological data analysis.
RealNeuralNetworks.jl
Due to the string-like nature of neurons and blood vessels, they could be abstracted as curved tubes with center lines and radii.