Florin Gogianu

@FlorinGogianu

Curious about machine learning, dabbled in philosophy of science. Code:

Vrijeme pridruživanja: listopad 2013.

Tweetovi

Blokirali ste korisnika/cu @FlorinGogianu

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

  1. Prikvačeni tweet
    4. srp 2019.

    So I made it through my first poster session at , presenting ways of sampling from the replay memory based on the uncertainty of the value function approximator.

    Poništi
  2. proslijedio/la je Tweet

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

    Poništi
  3. proslijedio/la je Tweet

    TCS Twitter, what are your absolute favorite textbooks?

    Poništi
  4. 1. sij

    Another call for the most interesting results in ML in 2019.

    Poništi
  5. 27. pro 2019.

    I highly recommend blog. Several times I found myself revisiting some of the posts knowing I’ll find certain concepts well explained in there.

    Poništi
  6. proslijedio/la je Tweet
    21. pro 2019.

    Have you heard of "RL as Inference"? ... you might be surprised that this framing completely ignores the role of uncertainty! (confusing, since it talks a lot about "posteriors") Our spotlight tries to make sense of this:

    Poništi
  7. 20. pro 2019.

    I've been wondering the same so I hope this gets many comments.

    Poništi
  8. proslijedio/la je Tweet
    20. pro 2019.

    We're pleased to let you know that your submission, On Bonus Based Exploration Methods In The Arcade Learning Environment, has been accepted at ! This huge endeavor was led by . W/ , & . More👇🏼

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet

    Our '19 Holiday Contest is live!🎁🌟💚 How to enter? 1. RT this photo. 2. Say something nice about a fellow gamer (& tag them) in the replies with 3. If selected, you + the friend you tagged BOTH win an RTX 2080 Ti! Learn more:

    Poništi
  10. proslijedio/la je Tweet
    4. pro 2019.

    Very excited about my new paper! We formulate the on-policy max-return RL objective w.r.t *arbitrary* offline data and without *any* explicit importance correction. Amazingly, the gradient of the objective w.r.t pi is exactly the on-policy policy gradient!

    Poništi
  11. proslijedio/la je Tweet
    4. pro 2019.

    I made a flowchart to help people choose which causal inference book to read :)

    Poništi
  12. 27. stu 2019.

    Check-out the work done by my colleagues on for learning interaction representations in videos.

    Poništi
  13. proslijedio/la je Tweet
    25. stu 2019.

    Causality for Machine Learning. (arXiv:1911.10500v1 [cs.LG])

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

    Book draft with proofs on RL

    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    21. stu 2019.

    Improving episodic control w/ Memory-Efficient Episodic Control Reinforcement Learning with Dynamic Online k-means: Sample-Efficient Reinforcement Learning with Maximum Entropy Mellowmax Episodic Control:

    Poništi
  16. proslijedio/la je Tweet
    13. stu 2019.

    Have you, like me, had a vague sense that work was happening to make model uncertainty more calibrated, but not known in more detail where the field was? This paper - which does broad testing of different uncertainty methods - is for you!

    Poništi
  17. proslijedio/la je Tweet
    4. stu 2019.

    Folks, have a winner. For the stochastic bandit, this gem from Odalric-Ambrym Maillard's thesis, called EwS, works incredibly well and gives explicit probabilities on the arms. I compare EwS with Thompson & Tsallis in

    Prikaži ovu nit
    Poništi
  18. proslijedio/la je Tweet
    5. stu 2019.

    We are hiring in our approx-Bayes team at RIKEN-AIP. Post-docs (2 positions), RA (4 positions), interns (15 positions). Job posting: Team page: Email: jobs-abi-riken-aip@googlegroups.com Help me spread the word. Retweets appreciated

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet

    School segregation in Romania. Assigning Roma children to different buildings, scheduling break times at different hours to avoid interaction with white kids. Just in case you thought the recent news cycle about Bulgarian racist laws was isolated.

    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    9. lis 2019.

    Thank you ! For those interested, our CS234 reinforcement learning lecture notes and assignments are also available here:

    Poništi
  21. proslijedio/la je Tweet
    25. ruj 2019.

    Excited to share that I’m teaching a *new course* on multi-task & meta-learning! Topics incl. optimization-based meta-learning, lifelong learning, meta-RL
, etc Slides & assignments being posted. Lecture videos to be publicly released after the course.

    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:

    ·