Dimensionality Reduction on Neural Data

Exploring dimensionality reduction with classification of neural data.

The original dataset and article Sofroniew, Nicholas James et al. “Neural coding in barrel cortex during whisker-guided locomotion.” can be found on the author's GitHub repo.

Contents

• PCA applied to a waveform to cell type problem
• An autoencoder applied to a waveform to cell type problem

Other references:

Cunningham, J., Yu, B. Dimensionality reduction for large-scale neural recordings. Nat Neurosci 17, 1500–1509 (2014). https://doi.org/10.1038/nn.3776

Paninski L, Cunningham JP. Neural data science: accelerating the experiment-analysis- theory cycle in large-scale neuroscience. Curr Opin Neurobiol. 2018 Jun;50:232-241. doi: 10.1016/j.conb.2018.04.007. PMID: 29738986.

Wu, Tong et al. “Deep Compressive Autoencoder for Action Potential Compression in Large-Scale Neural Recording.” Journal of Neural Engineering 15.6 (2018): n. pag. Journal of Neural Engineering. Web.

Ladjal, Saïd, Alasdair Newson, and Chi Hieu Pham. “A PCA-like Autoencoder.” arXiv 2 Apr. 2019: n. pag. Print.

Scree and cumulative explained variance plots

Matplotlib 3D scatter plot

Keras autoencoder guide

Hyperparameter grid search for Keras:

• https://machinelearningmastery.com/grid-search-hyperparameters-deep-learning-models-python-keras/
• https://stackoverflow.com/questions/49823192/autoencoder-gridsearch-hyperparameter-tuning-keras
• https://towardsdatascience.com/autoencoders-for-the-compression-of-stock-market-data-28e8c1a2da3e