Bayesian Methods Research Group

@bayesgroup

Research in Bayesian Deep Learning, Reinforcement Learning, Optimization, Structured Prediction, Drug Discovery and more. From Russia with ❤ and

Vrijeme pridruživanja: srpanj 2017.

Tweetovi

Blokirali ste korisnika/cu @bayesgroup

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @bayesgroup

  1. More from our research seminar: talks about current state of uncertainty estimation in Deep Learning

    Poništi
  2. Low-variance Gradient Estimates for the Plackett-Luce Distribution by , and Dmitry Vetrov in collaboration with Christopher Robinson and Novi Quadrianto

    Prikaži ovu nit
    Poništi
  3. Unsupervised Domain Adaptation with Shared Latent Dynamics for Reinforcement Learning by , and Dmitry Vetrov

    Prikaži ovu nit
    Poništi
  4. Pitfalls of In-Domain Uncertainty Estimation and Ensembling in Deep Learning by , Alexander Lyzhov, and Dmitry Vetrov

    Prikaži ovu nit
    Poništi
  5. It's workshops day at ! Check out our recent results at the Bayesian Deep Learning workshop (West Exhibition Hall C) Poster spotlights at 9:15, poster session at 9:30

    Prikaži ovu nit
    Poništi
  6. This evening and will present their work "The Implicit Metropolis-Hastings Algorithm" 05:00 -- 07:00 PM @ East Exhibition Hall B + C #183

    Poništi
  7. 3. Importance Weighted Hierarchical Variational Inference by 05:30 -- 07:30 PM @ East Exhibition Hall B + C #167

    Prikaži ovu nit
    Poništi
  8. 2. A Simple Baseline for Bayesian Uncertainty in Deep Learning by our alumni in collaboration with group 10:45 AM -- 12:45 PM @ East Exhibition Hall B + C #146

    Prikaži ovu nit
    Poništi
  9. 1. A Prior of a Googol Gaussians: a Tensor Ring Induced Prior for Generative Models by and 10:45 AM -- 12:45 PM @ East Exhibition Hall B + C #119

    Prikaži ovu nit
    Poništi
  10. BayesGroup at ! Come see our posters! Today we'll be presenting...

    Prikaži ovu nit
    Poništi
  11. By , , and Dmitry Vetrov in collaboration with Christopher Robinson and Novi Quadrianto. The paper was also accepted as a spotlight to the Bayesian Deep Learning workshop.

    Prikaži ovu nit
    Poništi
  12. Check out our new paper "Low-variance Black-box Gradient Estimates for the Plackett-Luce Distribution", accepted as an oral to () on how to reduce the variance of gradients when optimizing w.r.t. a distribution over permutations. Paper:

    Prikaži ovu nit
    Poništi
  13. Recently, Andrey Malinin of Research visited our research seminar to present his work on "Reverse KL-Divergence training of Prior Networks" ()

    Poništi
  14. proslijedio/la je Tweet
    5. stu 2019.

    Invited speaker: Anton Osokin () - a Leading Research Fellow at the Samsung-HSE Laboratory, Moscow, Russia, where he works with the . He was a post-doc at the SIERRA lab and then at the WILLOW lab at INRIA/ENS in Paris.

    Prikaži ovu nit
    Poništi
  15. Another paper is out! A new plug-and-play prior for generative models that efficiently packs over 10^100 Gaussians into a high dimensional grid.

    Poništi
  16. Next in our research seminars we had a guest talk by Eugene Golikov on theoretical understanding of Deep Learning

    Poništi
  17. Maxim Kodryan gave a talk on the "Invariant Risk Minimization" paper (by Martín Arjovsky)

    Poništi
  18. Next on our research seminar Sergey Troshin gave a talk covering the "Deep Equilibrium Models" paper (by )

    Poništi
  19. New academic year, new research seminars! In the first talk, gave an overview of current state of the Mutual Information estimation in Machine Learning.

    Poništi
  20. A good scientific work is not only in discovering novel results, but also in providing critical assessments to peers. We're proud that , , , , Dmitry Vetrov and Dmitry Kropotov were recognized as top 50% reviewers!

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·