Tejas Kulkarni

@tejasdkulkarni

Understanding our minds and brains to create AI. Scientist at Google DeepMind. Previously PhD @ MIT Brain&CogSci.

Vrijeme pridruživanja: srpanj 2009.
29 fotografija ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @tejasdkulkarni

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

  1. 2. sij

    Neuron as a computer

    Brains are amazing. Our lab demonstrates that single human layer 2/3 neurons can compute the XOR operation. Never seen before in any neuron in any other species. Out now in . Congrats Albert, Tim  & CO
    17 korisnika označava da im se sviđa
    Poništi
  2. 31. pro 2019.

    Let’s hope that in 2020 more people/voices in AI admit to not knowing. This should open up the space for new ideas and expressions

    1 reply 10 proslijeđenih tweetova 107 korisnika označava da im se sviđa
    Poništi
  3. 17. pro 2019.

    20 korisnika označava da im se sviđa
    Poništi
  4. 10. pro 2019.

    I will be at the to present this work tomorrow and more (Wednesday, 10:45 AM -- 12:45 PM @ East Exhibition Hall B + C. Poster #124). Come say hello if you are around!

    0:30
    Deep RL agents are data hungry and often learn task-specific representations. Our model learns object-centric abstractions from raw videos. This enables highly data-efficient RL and structured exploration.
    7 proslijeđenih tweetova 56 korisnika označava da im se sviđa
    Poništi
  5. proslijedio/la je Tweet
    20. stu 2019.

    MIT news is featuring our work on AI agents that learn to find friends and foes in a multi-agent game. To be presented at as a spotlight talk.

    3 proslijeđena tweeta 27 korisnika označava da im se sviđa
    Poništi
  6. proslijedio/la je Tweet
    30. lis 2019.

    DeepMind’s AlphaStar beat all but the very best humans at the fast-paced, strategy sci-fi game StarCraft II.

    19 replies 304 proslijeđena tweeta 817 korisnika označava da im se sviđa
    Poništi
  7. 24. lis 2019.

    We have released the code for transporter — a neural network architecture for unsupervised learning of object keypoints (now a NeurIPS paper):

    0:30
    Deep RL agents are data hungry and often learn task-specific representations. Our model learns object-centric abstractions from raw videos. This enables highly data-efficient RL and structured exploration.
    1 reply 58 proslijeđenih tweetova 290 korisnika označava da im se sviđa
    Poništi
  8. 20. lis 2019.

    12 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  9. 20. lis 2019.

    Double rainbow from the balcony 😌

    1 reply 24 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    10. lis 2019.

    Wanna play around with SPIRAL but the installation seems complicated? I've just built a Docker image to make the experience as hassle-free as possible. To get the agent up and running on your machine follow the instructions here: Have fun!

    0:07
    The 19-stroke SPIRAL agent for the oil paint simulator is now available on ! Enjoy 🎨🖌🤖 For usage instructions see The pre-trained snapshot is here:
    44 proslijeđena tweeta 175 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    8. lis 2019.

    We are excited to release our work on DexPilot, a markerless, glove-free and vision based teleoperation of dexterous robot hand-arm system pdf is here link to more videos

    13 replies 89 proslijeđenih tweetova 291 korisnik označava da mu se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    4. lis 2019.

    The 19-stroke SPIRAL agent for the oil paint simulator is now available on ! Enjoy 🎨🖌🤖 For usage instructions see The pre-trained snapshot is here:

    95 proslijeđenih tweetova 393 korisnika označavaju da im se sviđa
    Poništi
  13. proslijedio/la je Tweet
    3. lis 2019.

    The point is that these agents' representations can actually be instantiated in physical reality. This is not necessarily the case with purely neural autoencoder representations. Read more about the video below here:

    1 reply 5 proslijeđenih tweetova 20 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet
    3. lis 2019.

    Now for something different! Deep RL + GAN training + CelebA = artificial caricature. Agents learn to draw simplified (artistic?) portraits via trial and error. @ creativity workshop. Animated paper: PDF: Thread.

    212 proslijeđenih tweetova 618 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    25. ruj 2019.

    Congratulations Josh Tenenbaum on the “genius grant” — so lucky to have had you as my thesis advisor! Richly deserved

    4 proslijeđena tweeta 90 korisnika označava da im se sviđa
    Poništi
  16. 18. ruj 2019.

    The main claim about multi agent vs intrinsic motivation is not well supported. If you want to incentivize exploration so that objects interact in meaningful ways then explore with object/relation oriented goals, rather than distilling search

    We've observed AIs discovering complex tool use while competing in a simple game of hide-and-seek. They develop a series of six distinct strategies and counterstrategies, ultimately using tools in the environment to break our simulated physics:
    Prikaži ovu nit
    1 proslijeđeni tweet 16 korisnika označava da im se sviđa
    Poništi
  17. 3. ruj 2019.

    Interesting article on what it would take to plant 1 trillion trees in order to absorb up to 20 years worth of greenhouse gas emissions. I am not sure if there is a technological solution but maybe robots/drones/computer-vision could help accelerate?

    5 proslijeđenih tweetova 25 korisnika označava da im se sviđa
    Poništi
  18. proslijedio/la je Tweet
    23. kol 2019.

    NASA maps carbon monoxide from fires from orbit:

    169 replies 3.463 proslijeđena tweeta 5.862 korisnika označavaju da im se sviđa
    Poništi
  19. proslijedio/la je Tweet
    16. kol 2019.

    SPIRAL is an RL agent that generates images by interacting with off-the-shelf graphics software. We've released 9 agents trained to paint portraits of imaginary people in just 19 brush strokes on ! For more, see: → →

    1 reply 205 proslijeđenih tweetova 598 korisnika označava da im se sviđa
    Poništi
  20. proslijedio/la je Tweet
    13. kol 2019.

    Really excited to release to the public! - Clear, scalable experiments that test core capabilities. - Works with OpenAI gym, Dopamine. - Detailed colab analysis - Automated LaTeX appendix Example report:

    52 proslijeđena tweeta 203 korisnika označavaju da im se sviđa
    Poništi

@tejasdkulkarni još nije objavio nijedan Tweet.

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:

    ·