Pierre Glaser

@PierreGlaser

ML PhD Student in the UCL Gatsby Unit working on kernel methods and generative models.

London

pierreglaser.github.io

Joined August 2017

Pierre Glaser
@PierreGlaser
Jul 25, 2024
Happening now (11:30am-1pm)! Come chat with us at booth no. 1800 ⚙️
Pierre Glaser
@PierreGlaser
Jul 24, 2024
Given the energy spent improving predictive protein sequence models, can we find out how close they are to the ground truth? Log-likelihoods and RMSEs are no good for this, and can only be used for model comparison. (Deep) Kernel Methods to the rescue! (ICML 2024 Poster) 1/n
Pierre Glaser
@PierreGlaser
Jul 24, 2024
Given the energy spent improving predictive protein sequence models, can we find out how close they are to the ground truth? Log-likelihoods and RMSEs are no good for this, and can only be used for model comparison. (Deep) Kernel Methods to the rescue! (ICML 2024 Poster) 1/n
Pierre Glaser
@PierreGlaser
Aug 23, 2023
vmap-able distributions just dropped on numpyro 0.13! A (not so short) thread on what that means :D (1/11)
Pyro
@PyroAi
Aug 23, 2023
NumPyro 0.13.0 is released with several enhancements and bug fixes. Pytree Distribution has a new design that makes it work under jax.vmap, thanks to @PierreGlaser. github.com/pyro-ppl/numpy…
Pierre Glaser
@PierreGlaser
Aug 9, 2023
New work alert: Fast calibration tests for unnormalized models, presented #UAI2023 last week! A thread:
Pierre Glaser
@PierreGlaser
Dec 10, 2021
If you're interested, please come chat with us about our #NeurIPS2021 work on relaxing the KL gradient flow with @MichaelArbel and @ArthurGretton during today's poster session (8:30am PST)! neurips.cc/virtual/2021/p…
GIF