Shashwat Sridhar

The paper develops a spatial contrast model to capture nonlinear spatial integration in primate retinal ganglion cells using natural and artificial stimuli, showing improved performance over linear models.

This study examines how primate retinal ganglion cells adapt their spatial processing to natural versus artificial stimuli, finding that some cell types adjust receptive fields and surrounds, but efficient coding only partially explains these changes.

This study uses convolutional neural networks to investigate wide-field inhibition in the marmoset retina, revealing hidden computational aspects of retinal processing through interpretable models.

This paper introduces STR, a method combining super-resolution and tomography to identify receptive field subunits in retinal ganglion cells, validated in primate recordings.

The paper introduces an accelerated version of spike-triggered non-negative matrix factorization to identify coordinated subunit mosaics in primate ganglion cells.

This study compares predictive models for retinal ganglion cell responses, finding that CNNs outperform LN models, while LN models show better cross-stimulus generalization.

Kilosort4 swaps K-means-style clustering for a bipartite modularity graph method with a merging tree, and recovers most ground-truth neurons in simulations where competing sorters lose half of them.