This work explores Liquid Time-Constant Networks (LTCs) and Closed-form Continuous-time Networks (CfCs) for modeling retinal ganglion cell activity in tiger salamanders across three datasets. Compared to a convolutional baseline and an LSTM, both architectures achieved lower MAE, faster convergence, smaller model sizes, and favorable query times, though with slightly lower Pearson correlation. Their efficiency and adaptability make them well suited for scenarios with limited data and frequent retraining, such as edge deployments in vision prosthetics.
@article{arxiv.2511.18014,
title = {Modeling Retinal Ganglion Cells with Neural Differential Equations},
author = {Kacper Dobek and Daniel Jankowski and Krzysztof Krawiec},
journal= {arXiv preprint arXiv:2511.18014},
year = {2025}
}
Comments
Accepted to the AAAI-26 Student Abstract and Poster Program, with supplementary material