Rezultati pretraživanja
  1. 30. lis 2019.

    Our work (SILO) addresses one challenge of imitation learning in robotics: agents/environments between a learner and a demonstrator are often different! SILO is a HRL method that selects the best imitable frames and finds a way to follow them.

    1 reply 15 proslijeđenih tweetova 63 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
  2. 1. stu 2019.

    CoRL Awards!

    1 reply 5 proslijeđenih tweetova 23 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
  3. 4. stu 2019.

    Having robot learning blues after the end of ? Fear not! Our new paper takes sim2real further and explores how we can learn to one-shot imitate humans (using TecNets) without any real-world data during training!

    1 reply 11 proslijeđenih tweetova 33 korisnika označavaju da im se sviđa
  4. 30. lis 2019.

    So many great sim-to-real techniques being presented at in Osaka 🇯🇵

    5 korisnika označava da im se sviđa
  5. 17. lis 2019.

    How do drivers predict future pedestrian behavior? Our map-aware model, the Discrete Residual Flow Network, predicts multimodal and uncertain pedestrian behaviors over long time horizons. We'll be giving a talk at ! Paper:

    8 proslijeđenih tweetova 15 korisnika označava da im se sviđa
  6. 4. stu 2019.

    🚨 New blog post! 🚨 Ben Abbatematteo discusses his paper "Learning to Generalize Kinematic Models to Novel Objects."

    1 reply 3 proslijeđena tweeta 5 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
  7. 1. stu 2019.

    George Konidaris delivered his keynote "Signal to Symbol (via Skills)"at . For those who missed it and want to hear about learning abstraction hierarchies and high-level symbolic representations, watch the talk here:

    1 reply 9 proslijeđenih tweetova 27 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
  8. 21. stu 2019.

    Cool single-actuator hand mechanism I saw at

    3 proslijeđena tweeta 23 korisnika označavaju da im se sviđa
  9. 12. stu 2019.

    paper from our Apple group on Worst Cases Policy Gradients: Learning more robust policies by minimizing long-tail risks, reducing the likelihoods of bad outcomes. with Charlie Tang and Jian Zhang.

    26 proslijeđenih tweetova 79 korisnika označava da im se sviđa
  10. 1. stu 2019.

    Learning by Cheating. Decoupling learning to see from learning to act. New cool thought-provoking work from Vladlen Koltun.

    22 proslijeđena tweeta 72 korisnika označavaju da im se sviđa
  11. 1. stu 2019.

    Congratulations to the 3rd it was an honor to give a keynote on to the So excited to share the common goals w this robotics community:

    ***Reserve your seats NOW*** Registration will be open at the following URL until Monday Sept. 30, 2019（JST）! [NOTE] We can offer 420 seats on a first come, first serve basis. We look forward to having you in Osaka!!!
    2 proslijeđena tweeta 17 korisnika označava da im se sviđa
  12. 30. lis 2019.

    Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learning () article arxiv

    New paper - RelayPL Takeaway: Goal conditioned hierarchical policy when infused with unstructured play data and a relabelling trick can solve REALLY LONG horizon tasks! Website: With A. Gupta, , ,
    Prikaži ovu nit
    9 proslijeđenih tweetova 38 korisnika označava da im se sviđa
  13. 30. lis 2019.

    At ? Check out "Active Domain Randomization" at poster 2C-02 in today's poster session. paper: . and will be presenting it.

    1 reply 5 proslijeđenih tweetova 21 korisnik označava da mu se sviđa
  14. 30. lis 2019.

    Check out the great talk on AC-TEACH by at Video:

    Our new work at will present RL with Ensemble of Suboptimal Teachers -aka- specify as much as you can easily, let learning handle the rest. Blog: Paper: w\ , A. Mandlekar, ,
    Prikaži ovu nit
    5 proslijeđenih tweetova 20 korisnika označava da im se sviđa
  15. 29. lis 2019.

    Introducing our work on hierarchical planning

    0:52
    We introduce CAVIN to hierarchically generate plans in learned latent spaces. Our robot learns to strategically interact with multiple objects in different tasks. Webpage: Paper: w/
    Prikaži ovu nit
    6 proslijeđenih tweetova 27 korisnika označava da im se sviđa
  16. 29. lis 2019.

    Preferred Networks is sponsoring which is held at Osaka from Oct.30 to Nov.1, 2019. We exhibit Chainer RL, Humanoid hand, and Invisible marker at our booth #2. Please come to check it!

    6 proslijeđenih tweetova 21 korisnik označava da mu se sviđa
  17. 29. lis 2019.

    【ブース展示】2019年10月30日(水)〜11月1日(金)まで、大阪千里ライフサイエンスセンターで開催される にPFNブースを出展しています。 ChainerRL、Humanoid hand、Invisible markerを展示していますので是非お越しください！

    5 proslijeđenih tweetova 13 korisnika označava da im se sviđa
  18. 29. lis 2019.

    I got to play with a really cool robot hand and see a demo with invisible ink marker (yet visible in UV) at HQ this week. They just told me they'll also have them at their sponsor booth at , should be fun to check out!

    1 reply 29 proslijeđenih tweetova 93 korisnika označavaju da im se sviđa
  19. 29. lis 2019.

    Folks at , check out our work w/ on better-than-demonstrator imitation without ranking info () and on the role of human gaze in imitation learning ().

    4 proslijeđena tweeta 29 korisnika označava da im se sviđa
  20. 29. lis 2019.

    お！！今年のCoRLの論文が全てアップロードされている！今年はレビューも公開されているらしい

    17 proslijeđenih tweetova 61 korisnik označava da mu 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.