An Overview of Probabilistic Latent Variable Models

Add to your list(s) Download to your calendar using vCal

• Aditya Ravuri (Computer Lab)
• Wednesday 14 February 2024, 10:00-11:00
• Martin Ryle Seminar Room, KICC.

If you have a question about this talk, please contact km723.

This talk showcases some interesting probabilistic interpretations of dimensionality reduction and unsupervised representation learning algorithms and presents the common statistical modelling assumptions that underpin these algorithms. Specifically, we’ll look at methods such as PCA , GMM, ICA , FA, VAE and GPLVM and show how these share the same modelling framework. If there’s interest, I’ll also talk about a large class of other algorithms that also fit into this framework (such as t-SNE, UMAP , isomap and MDS ). I’ll cover some newer methods like contrastive learning and show how these fit in with classical latent variable models.

This talk is part of the Astro Data Science Discussion Group series.

This talk is included in these lists:

• Astro Data Science Discussion Group
• Cambridge Astronomy Talks
• Combined External Astrophysics Talks DAMTP
• Cosmology, Astrophysics and General Relativity
• Institute of Astronomy Extra Talks
• Institute of Astronomy Talk Lists
• Martin Ryle Seminar Room, KICC

Note that ex-directory lists are not shown.