Rezultati pretraživanja
  1. Moving onto new textbook & state-space models for modelling class today!

  2. prije 9 sati
  3. 📣 new paper alarm 🙂 We provide a method to parametrize simulation models – The method automatically chooses: – summary statistics – weighs them – uses these to parametrize models without running direct minimization

  4. prije 14 sati

    Networks in Healthcare: Distribution by Medical Condition. (arXiv:2002.00224v1 [])

  5. prije 20 sati

    a tutorial on bayesian neural nets by wesley ( pretty decent collaborator ) (altho im not a bayesian ;)

  6. 3. velj

    Bayes rule (politics twitter version): Posterior = prior * prior / prior

  7. 3. velj

    Inference: How Grid Approximation Works by ___

  8. 3. velj

    So today in 🤦‍♂️moments I learned that runif is a function to generate random numbers with a uniform distribution, and should be read r uniform, not run if (which clues you in to what I thought it did..)

  9. 3. velj

    Massive thanks to my supervisors Bob Furness, Mark Trinder. And my examiners and Morten Frederiksen. PhD Viva passed . Watch this space...

  10. 3. velj

    Customer surveys are naturally prone to biases. One prominent example is participation bias. Here we tackle participation bias for the case of the net promoter survey by means of multilevel regression and poststratification.

  11. 2. velj
    Odgovor korisniku/ci

    Also have to consider published studies in Lancet that cite 11 and 14% death rates in small samples. It's all speculative but each data point gets us closer.

  12. 2. velj

    A Bigram Poem inspired by zevdatascience: A machine machine scientist scientist to to aid aid in in the -zeveigen

  13. 1. velj

    The Case for . (arXiv:2001.10995v1 [cs.LG])

  14. 1. velj
    Odgovor korisniku/ci

    I also did a Inversion and the source Parameters are well constrained by

  15. 1. velj
  16. 31. sij
  17. 31. sij

    Rockstar. Sailor woman. Bird woman. And now Doctor too. Well done for defending her PhD on , and stuff at !

  18. 31. sij

    Reasoning with Deep-Learned Knowledge. (arXiv:2001.11031v1 [cs.LG])

  19. 31. sij

    Reinforcement Learning via Deep, Sparse Sampling. (arXiv:1902.02661v3 [cs.LG] UPDATED)

  20. 30. sij

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.