I-7 (Mon 16:30): @SchulzAuguste kicks off the @mackelab poster marathon, introducing LDNS, a diffusion-based latent variable model to generate diverse neural spiking data flexibly conditioned on external variables.
Fun project co-led with @_Jaivardhan_: arxiv.org/abs/2407.08751View quoted post on X
We’re stoked to share: “A Practical Guide to Sample-based Statistical Distances for Evaluating Generative Models in Science”.
Now out in TMLR: openreview.net/forum?id=isEFz…
This was an incredibly special project for us, as it involved the **entire** lab getting together!View quoted post on X
@gloecklermanuel and @deismic_ are at ICML to present our work on `All-in-one simulation-based inference`. Join us at ORAL 6F tomorrow (Thursday) 4:45pm, or come by poster 1309, tomorrow 1:30pm!View quoted post on X
We are very excited to share our new work, which unifies probabilistic neural encoding, behavioral decoding, and joint dimensionality reduction in a single model.
How?
Through modeling conditional distributions in VAEs!
Preprint: biorxiv.org/content/10.110…View quoted post on X
