Rezultati pretraživanja
  1. 1. lis 2019.

    The RecoGym Challenge 100 is Live:

    1 reply 23 proslijeđena tweeta 33 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
  2. 25. ruj 2019.

    Slides from my talk at the workshop on "Recap: Designing a more Efficient Estimator for Off-policy Evaluation in Bandits with Large Action Spaces"

    13 proslijeđenih tweetova 39 korisnika označava da im se sviđa
  3. 25. ruj 2019.

    Slides of my REVEAL workshop talk "Engagement, metrics and recommenders". Following from focus is on recommendations, I see connections (not established yet!) between engagement, long vs short and reinforcement learning (rewards).

    17 proslijeđenih tweetova 61 korisnik označava da mu se sviđa
  4. 24. ruj 2019.

    The proceedings of workshop on Interfaces and Human Decision Making for Recommender Systems - are now available online:

    14 proslijeđenih tweetova 29 korisnika označava da im se sviđa
  5. 24. ruj 2019.

    This year was biggest edition to date, we had a blast being among the 909 attendees! This year we could see that the field is maturing, as we tell you in our last blog article by Ugo and Pranjul

    16 proslijeđenih tweetova 39 korisnika označava da im se sviđa
  6. 22. ruj 2019.

    Been following the or hashtags recently? Did you know maintains a dates and deadline calendar for all 24 conferences we sponsor? Upcoming list is here:

    1 reply 11 proslijeđenih tweetova 31 korisnik označava da mu se sviđa
  7. 21. ruj 2019.

    Gave the last talk on "Sequence aware Reinforcement Learning on Knowledge Graphs" in the last session of the workshop yesterday! We propose leveraging item sequence information when training RL models over product knowledge graph work w/

    Image shows a projector screen with a slide showing that knowledge graphs are useful
    The image shows a person standing in front of a slide which mentions details of a sequence aware policy / value network
    1 reply 9 proslijeđenih tweetova 38 korisnika označava da im se sviđa
  8. 20. ruj 2019.

    Impossible to choose a single picture to capture the many, many highlights of a great ! The sense of community remains alive even in a sea of 900 attendees! Looking forward to in Rio!

    5 proslijeđenih tweetova 29 korisnika označava da im se sviđa
  9. 20. ruj 2019.

    Invited talk on metrics/engagement and recommender systems by from at workshop at

    6 proslijeđenih tweetova 26 korisnika označava da im se sviđa
  10. 19. ruj 2019.

    ＼国際学会 AI Labから3件のポスター発表中！／ コペンハーゲンで開催されている は、情報推薦，推薦システムに関する国際学会です。 16日には産学連携を行っている、 京都大学 鹿島・山田研究室&リサーチインターン生と共に取組んだ研究についてポスター発表がありました！

    1 reply 15 proslijeđenih tweetova 35 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
  11. 19. ruj 2019.

    If you're at and like bandits, you should definitely come to tomorrow! Both our papers can now be found on ArXiv: and . , , David Rohde, , , Alexandre Gilotte.

    14 proslijeđenih tweetova 24 korisnika označavaju da im se sviđa
  12. 19. ruj 2019.

    On my way to INRA workshop to present results of great teamwork sanne Vrijenhoek on democratic diversity metrics

    6 proslijeđenih tweetova 18 korisnika označava da im se sviđa
  13. 18. ruj 2019.

    Curious about what our team does to personalize the Home page of Spotify? Here's what Oguz Semerci recently presented at , which gives an overview of how we rank content on Home, and shares some lessons learned in sanity checking our ML systems

    19 proslijeđenih tweetova 53 korisnika označavaju da im se sviđa
  14. 17. ruj 2019.

    Very disappointing to see a paper describing Youtube's multi-objective , but failing to acknowledge there should be ranking objectives besides pure user engagement or more important biases than "positional" bias

    6 replies 49 proslijeđenih tweetova 209 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
  15. 17. ruj 2019.

    groupon、推薦理由を58種のメッセージから選択して表示するテストをして、50%以上の改善を実現。　

    12 proslijeđenih tweetova 51 korisnik označava da mu se sviđa
  16. 17. ruj 2019.

    Spotify 💚RecSys! Come talk to us about the work we're doing on recommender systems and personalization

    10 proslijeđenih tweetova 30 korisnika označava da im se sviđa
  17. 16. ruj 2019.

    The team placed 15 out of 1534 in the Challenge. Learn how we used RAPIDS and PyTorch to speed up 's tabular model by 15X!

    Excited to be at sharing work on how we used and to accelerate our recsys challenge submission by 15.6x! Lots of great deep learning optimizations and tricks in the repo. Check out the blog for details:
    15 proslijeđenih tweetova 23 korisnika označavaju da im se sviđa
  18. 16. ruj 2019.

    The slides of my opening keynote are now available

    27 proslijeđenih tweetova 77 korisnika označava da im se sviđa
  19. 15. ruj 2019.

    Off to Ping me if you want to talk with me about this position :)

    11 proslijeđenih tweetova 15 korisnika označava da im se sviđa
  20. 14. ruj 2019.

    Heading to Copenhagen for . My team has two long papers: Neural Deep Retrieval for dealing with large item catalogs; Multi-task ranking for doing multi-objective optimization. Both papers contain lessons from dealing with large scale recommendation problems at Google.

    13 proslijeđenih tweetova 79 korisnika označava da im se sviđa

Nema rezultata.

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.