NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:9406
Title:Enabling hyperparameter optimization in sequential autoencoders for spiking neural data

The paper demonstrates that the sequential autoencoder that's becoming popular in neuroscience is prone to overfitting and propose solutions to address this overfitting. It is overall a good paper. A couple of small issues could be corrected to improve the quality: - Baselines such as dAEs that were in the feedback should be put in the main paper along with running baselines on the real datasets - A discussion of alternatives and baselines to motivate the solution