Tweetovi

Blokirali ste korisnika/cu @BenBeyret

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

  1. Prikvačeni tweet
    30. tra 2019.

    The day has arrived! The Animal-AI Olympics training environment is out. Check out the repo for more details: with

    Poništi
  2. proslijedio/la je Tweet
    2. velj

    A story in three acts

    Prikaži ovu nit
    Poništi
  3. 10. sij

    This is infuriating 😠 For our competition, I spent days considering how malicious participants might game the process. It's hard, and I still find ways things could have gone wrong today Sadly, the best solution was to restrict user inputs and blackboxing the evaluation

    Poništi
  4. proslijedio/la je Tweet
    5. sij
    Odgovor korisniku/ci

    There's at least a growing breadth of interest in envs, such as: procedurally-generated , cognition-focused Animal-AI , embodiment in , and 's AI-GAs. And in intrinsic rewards: learning progress, empowerment, homeostasis, etc.

    Poništi
  5. proslijedio/la je Tweet
    19. pro 2019.

    We are looking for outstanding PhD applicants for next year. If want to explore how reinforcement learning, in particular Quality-Diversity algorithms, can make legged robots more adaptive, resilient and versatile, consider applying to my lab! More here

    Prikaži ovu nit
    Poništi
  6. proslijedio/la je Tweet
    14. pro 2019.

    And... another one. Curious about how to combine Hierarchical RL and Grammar Induction to define a powerful Action Grammar? Checkout my work with at the Deep RL & Transferable Skills Workshops. Preprint:

    Poništi
  7. 14. pro 2019.

    Following a light lunch, David Ha () talks "Innate Everything" Learning Transferable Skills workshop

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

    Great presentation of the by , , & the PPO-LSTM-based winning solution by . Very much looking forward to the next generation including human & animal expert traces!

    Poništi
  9. proslijedio/la je Tweet
    13. pro 2019.

    Join us tomorrow for an action-packed day with a fantastic line-up of speakers including , Vladlen Koltun, , , , and .

    Poništi
  10. 12. pro 2019.

    Looking forward to seeing everyone there, and my apologies to all the authors whose Twitter handle I couldn’t find Organisers: and myself!

    Prikaži ovu nit
    Poništi
  11. 12. pro 2019.

    "Multi-Task Reinforcement Learning without Interference" Tianhe Yu, , , , Saurabh Kumar and Abhishek Gupta "Efficient Exploration via State Marginal Matching" Lisa Lee, Benjamin Eysenbach, Emilio Parisotto, Eric Xing, ,

    Prikaži ovu nit
    Poništi
  12. 12. pro 2019.

    "Butterfly: A Panacea for All Difficulties in Wildly Unsupervised Domain Adaptation" Feng Liu,Jie Lu, Bo Han,Gang Niu, Guangquan Zhang, Masashi Sugiyama "Meta-Learning using Probabilistic Context Variables with Structured Latent Space" Hongyu Ren, ,

    Prikaži ovu nit
    Poništi
  13. 12. pro 2019.

    "Hallucinative Topological Memory for Visual Robotic Manipulation" , , Christine Tung, , Aviv Tamar "Soft Option Transfer" , , Matthew Smith, Wendelin Boehmer,

    Prikaži ovu nit
    Poništi
  14. 12. pro 2019.

    "ADAIL: Adaptive Adversarial Imitation Learning" Yiren Lu, Jonathan Tompson "Options of Interest: Temporal Abstraction with Interest Functions" Khimya Khetarpal, , , , Doina Precup

    Prikaži ovu nit
    Poništi
  15. 12. pro 2019.

    “Learning to Plan in High Dimensions via Neural Exploration-Exploitation Trees”, Binghong Chen, Bo Dai, Le Song “Semantic RL with Action Grammars: Data-Efficient Learning of Hierarchical Task Abstractions”,

    Prikaži ovu nit
    Poništi
  16. 12. pro 2019.

    And of course a poster session at 3:15pm: “Sparse Transfer Learning via Winning Lottery Tickets” “Zero-shot task adaptation by homoiconic meta-mapping” , James McClelland

    Prikaži ovu nit
    Poništi
  17. 12. pro 2019.

    2:55pm “Learning to Control Self-Assembling Morphologies: A Study of Generalization via Modularity” ,  Chris Lu, Trevor Darrell, Phillip Isola, Alexei Efros

    Prikaži ovu nit
    Poništi
  18. 12. pro 2019.

    2:35pm “Self-Attentional Credit Assignment for Transfer in Reinforcement Learning” by , Raphaël Marinier, Matthieu Geist, Olivier Pietquin

    Prikaži ovu nit
    Poništi
  19. 12. pro 2019.

    we have 3 great spotlights for accepted papers:

 2:15pm “Disentangled Skill Embeddings for Reinforcement Learning” by @janith_pet, Sergio Pascual-Diaz, Vincent Adam, Peter Vrancx and Jordi Grau-Moya

    Prikaži ovu nit
    Poništi
  20. 12. pro 2019.

    4:15pm Katja Hofmann () from on “Multi-task RL and generalisation" 5pm Wojciech Zaremba () from will conclude with his work on "Solving a Rubik’s cube using a robotic hand"

    Prikaži ovu nit
    Poništi
  21. 12. pro 2019.

    1:30pm David Ha () from will present “Innate Structures, Innate Brains, Innate World Models” 🐠

    Prikaži ovu nit
    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:

    ·