SeminarPast EventMachine Learning

Finding needles in the neural haystack: unsupervised analyses of noisy data

Wednesday, December 1, 2021
16:00 UTC
Marine Schimel & Kris Jensen

University of Cambridge, Department of Engineering

Hosted by CamBRAIN Virtual Journal Club70 minutes

Abstract

In modern neuroscience, we often want to extract information from recordings of many neurons in the brain. Unfortunately, the activity of individual neurons is very noisy, making it difficult to relate to cognition and behavior. Thankfully, we can use the correlations across time and neurons to denoise the data we record. In particular, using recent advances in machine learning, we can build models which harness this structure in the data to extract more interpretable signals. In this talk, we present two such methods as well as examples of how they can help us gain further insights into the neural underpinnings of behavior.

We use essential cookies to run the site. Analytics cookies are optional and help us improve World Wide. Learn more.