Summary of Diffusion-based Generation Of Neural Activity From Disentangled Latent Codes, by Jonathan D. Mccart et al.
Diffusion-Based Generation of Neural Activity from Disentangled Latent Codes
by Jonathan D. McCart, Andrew R. Sedler, Christopher Versteeg, Domenick Mifsud, Mattia Rigotti-Thompson, Chethan Pandarinath
First submitted to arxiv on: 30 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Neurons and Cognition (q-bio.NC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed approach leverages conditional generative modeling and InfoDiffusion to enable unsupervised inference of disentangled behavioral variables from recorded neural activity. The model, Generating Neural Observations Conditioned on Codes with High Information (GNOCCHI), is applied to time series neural data and compared to a VAE-based sequential autoencoder. GNOCCHI learns higher-quality latent spaces that are more structured and disentangled, enabling accurate generation of novel samples through simple linear traversal. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Scientists have developed new ways to record brain activity from thousands of neurons at once. This has led to the development of new methods for analyzing this data. The proposed method uses a type of generative model to find patterns in the data that can be used to predict what would happen if certain conditions were different. This could help researchers understand how the brain works and make predictions about behavior. The method is tested on real and simulated data from reaching tasks. |
Keywords
* Artificial intelligence * Autoencoder * Generative model * Inference * Time series * Unsupervised
