New theory paper on arXiv

Dec 5, 2023

We are excited to see our latest paper up on arXiv.

Doran K, Seifert M, Yovanovich CAM, Baden T§. Distance function for spike prediction. doi: https://doi.org/10.48550/arXiv.2312.01966. arXiv. direct link. pdf.
In this work, we present a new way to think about spike prediction based on a so called ‘distance function’:
Approaches to predicting neuronal spike responses commonly use a Poisson learning objective. This objective quantizes responses into spike counts within a fixed summation interval, typically on the order of 10 to 100 milliseconds in duration; however, neuronal responses are often time accurate down to a few milliseconds, and at these timescales, Poisson models typically perform poorly. To overcome this limitation, we propose the concept of a spike distance function that maps points in time to the temporal distance to the nearest spike. We show that neural networks can be trained to approximate spike distance functions, and we present an efficient algorithm for inferring spike trains from the outputs of these models. Using recordings of chicken and frog retinal ganglion cells responding to visual stimuli, we compare the performance of our approach to Poisson models trained with various summation intervals. We show that our approach outperforms the standard Poisson approach at spike train inference.