# Signal Processing in the Retina: Interpretable Graph Classifier to Predict Ganglion Cell Responses

arXiv: https://arxiv.org/abs/2401.01813 · HTML: https://mlsanity.com/paper/2401.01813/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 50 references that are in the map.
- First posted on arXiv 3 Jan 2024.

## Builds on (most cited first, 12 of 12)

- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [Graph Signal Processing: Overview, Challenges and Applications](https://mlsanity.com/paper/1712.00468/) (2017)
- [Faster Dynamic Matrix Inverse for Faster LPs](https://mlsanity.com/paper/2004.07470/) (2020)
- [Graph Spectral Image Processing](https://mlsanity.com/paper/1801.04749/) (2018)
- [Deep Learning Models of the Retinal Response to Natural Scenes](https://mlsanity.com/paper/1702.01825/) (2017)
- [PyTorchVideo: A Deep Learning Library for Video Understanding](https://mlsanity.com/paper/2111.09887/) (2021)
- [Feature Graph Learning for 3D Point Cloud Denoising](https://mlsanity.com/paper/1907.09138/) (2019)
- [Signed Graph Metric Learning via Gershgorin Disc Perfect Alignment](https://mlsanity.com/paper/2006.08816/) (2020)
- [From deep learning to mechanistic understanding in neuroscience: the structure of retinal prediction](https://mlsanity.com/paper/1912.06207/) (2019)
- [Searching for collective behavior in a network of real neurons](https://mlsanity.com/paper/1306.3061/) (2013)
- [A Spatiotemporal Oriented Energy Network for Dynamic Texture Recognition](https://mlsanity.com/paper/1708.06690/) (2017)

## Built on by (most cited first, 1 of 1)

- [Sparse Graph Learning from Sparse Data via Fiedler Number Maximization](https://mlsanity.com/paper/2604.26132/) (2026)

## Abstract (arXiv)

It is a popular hypothesis in neuroscience that ganglion cells in the retina are activated by selectively detecting visual features in an observed scene. While ganglion cell firings can be predicted via data-trained deep neural nets, the networks remain indecipherable, thus providing little understanding of the cells' underlying operations. To extract knowledge from the cell firings, in this paper we learn an interpretable graph-based classifier from data to predict the firings of ganglion cells in response to visual stimuli. Specifically, we learn a positive semi-definite (PSD) metric matrix M succeq 0 that defines Mahalanobis distances between graph nodes (visual events) endowed with pre-computed feature vectors; the computed inter-node distances lead to edge weights and a combinatorial graph that is amenable to binary classification. Mathematically, we define the objective of metric matrix M optimization using a graph adaptation of large margin nearest neighbor (LMNN), which is rewritten as a semi-definite programming (SDP) problem. We solve it efficiently via a fast approximation called Gershgorin disc perfect alignment (GDPA) linearization. The learned metric matrix M provides interpretability: important features are identified along M's diagonal, and their mutual relationships are inferred from off-diagonal terms. Our fast metric learning framework can be applied to other biological systems with pre-chosen features that require interpretation.