olivlogol

@olivlogol

Robotics, autonomous systems, IOT, IA and machine learning enthusiast. Software architect, developer and integrator for complex systems.

BZH
Vrijeme pridruživanja: siječanj 2010.

Tweetovi

Blokirali ste korisnika/cu @olivlogol

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

  1. proslijedio/la je Tweet
    3. velj

    We are delighted to announce two new innovative 🇪🇺 leaders in our group. Warm welcome to and and look forward to working with you✌🏻. Cc

    , , i još njih 5
    Poništi
  2. proslijedio/la je Tweet
    30. sij

    Humans learn from curriculum since birth. We can learn complicated math problems because we have accumulated enough prior knowledge. This could be true for training a ML/RL model as well. Let see how curriculum can help an RL agent learn:

    Poništi
  3. proslijedio/la je Tweet
    28. sij

    [FIC J-J ⌛️] L'équipe Silicom vous attend sur le stand D30 pour échanger sur ses nouvelles avancées ! A tout de suite !

    Poništi
  4. proslijedio/la je Tweet
    20. sij

    Intrinsically Motivated Open-Ended Learning in Autonomous Robots The special issue in the Frontiers in Neurorobotics journal is out! Intrinsically Motivated Open-Ended Learning in Autonomous Robots

    Prikaži ovu nit
    Poništi
  5. 16. sij

    Selon moi, c'est de ne pas suffisamment observer les formes qui conduit ces IA basées sur les réseaux convolutionnels à se tromper si lourdement via @scienceetvie

    Poništi
  6. 17. pro 2019.

    Très intéressant résumé d'un papier dans Nature traitant de notre comportement ilpulsif face à la nourriture. Où il est question d'apprentissage, de motivation, de renforcement, etc

    Poništi
  7. 12. pro 2019.

    Looking forward to reading the paper about from . Dreamer is capable of predicting the future in very few iterations. Look at boxing Atari game for example

    Poništi
  8. 10. pro 2019.

    In parallel with his talk at , interesting interview of Yoshua Bengio about the future of AI. Consciousness is one of the key challenges

    Poništi
  9. 1. pro 2019.

    A must read about consciousness and AI. Really fascinating work by and his team

    Poništi
  10. proslijedio/la je Tweet
    28. stu 2019.

    Elle sera l'unique école universitaire de recherche en en France: dès la rentrée 2020, proposera à une formation innovante et internationale. Objectif: doubler le nombre de diplômés rennais d’ici 10 ans, via

    , , i još njih 7
    Poništi
  11. proslijedio/la je Tweet
    26. stu 2019.

    We are happy to announce the v2.0 release of the Google Research Football Environment. The most exciting feature of this release is the Game Server, which lets your agent compete online with other researchers' models. Visit and give it a try!

    Poništi
  12. proslijedio/la je Tweet
    24. stu 2019.

    A key dimension of natural intelligence is in choosing what and when to learn (and whether the learning is done through curve fitting or something else is orthogonal). Beyond statistical learning, AI require advanced motivational/teleological architectures

    Poništi
  13. 15. stu 2019.
    Poništi
  14. proslijedio/la je Tweet
    14. stu 2019.

    We have written about our hypothesis on possible links between functions of consciousness and artificial general intelligence (in Japanese). We will enrich the contents and then translate it into a full paper in English. via

    Poništi
  15. 7. stu 2019.

    A must read this year :-) A really great invitation to think about best AI architectures

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

    Cool experimentation of modular intrinsically motivated goal exploration with a many-dofs hand that learns how to manipulate a pen. Could next steps of the agent's discoveries be drawing or pen flipping?

    Poništi
  17. proslijedio/la je Tweet
    29. lis 2019.

    Learning to Predict Without Looking Ahead: World Models Without Forward Prediction Rather than hardcoding forward prediction, we try to get agents to *learn* that they need to predict the future. Check out our paper!

    Prikaži ovu nit
    Poništi
  18. proslijedio/la je Tweet
    24. lis 2019.

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

    Poništi
  19. proslijedio/la je Tweet

    Rats generalize better than deep RL, and are more sample efficient

    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet

    I just published An Information Theory of Consciousness and Motivation

    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:

    ·