Suraj Narayanan S

@surajx32

Research Scientist . Tweets about Reinforcement Learning, and AI safety.

Vrijeme pridruživanja: kolovoz 2010.

Tweetovi

Blokirali ste korisnika/cu @surajx32

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @surajx32

  1. 20. stu 2019.

    I can't highlight this enough. A well executed and documented research work that resulted in a negative result often shed more light than a successful incremental work, especially in AI. Major conferences should actively promote submission of such papers.

    Poništi
  2. 29. lis 2019.

    Recommend reading for anyone starting out with TF2. Really well written.

    Poništi
  3. proslijedio/la je Tweet
    11. lis 2019.
    Poništi
  4. 15. velj 2019.

    It's not just writing, every bit of your data exhaust is contributing to the making of your future digital self.

    Poništi
  5. 14. velj 2019.

    This could pass off as the ramblings of a misinformed human. 😣

    Poništi
  6. proslijedio/la je Tweet
    12. velj 2019.

    Excited about ? Let's use it for cooperative multi-agent RL! Introducing : The StarCraft Multi-Agent Challenge, a diverse set of decentralised unit-level micro challenges. Preprint: Blog: Code:

    Prikaži ovu nit
    Poništi
  7. 2. velj 2019.

    Such an awesome initiative! Very useful, especially for baselining.

    Poništi
  8. proslijedio/la je Tweet

    Proteins are essential to life. Predicting their 3D structure is a major unsolved challenge in biology and could impact disease understanding and drug discovery. I’m excited to announce that we have won the CASP13 protein folding competition!

    Prikaži ovu nit
    Poništi
  9. 28. ruj 2018.

    “Building safe artificial intelligence: specification, robustness, and assurance”

    Poništi
  10. 20. ruj 2018.
    Poništi
  11. 20. ruj 2018.

    Brace yourself with this excellent visualisation of the Riemann zeta function.

    Poništi
  12. 19. ruj 2018.
    Poništi
  13. 15. ruj 2018.
    Poništi
  14. proslijedio/la je Tweet
    14. ruj 2018.

    Excited to share this paper. We generalize the momentum method for optimization, and in the process find that you can greatly extend the class of convex functions on which exponential convergence is possible using just gradients: Hamiltonian descent ().

    Prikaži ovu nit
    Poništi
  15. 12. ruj 2018.

    An illustrative introduction to the core concepts behind Tensorflow.

    Poništi
  16. 12. ruj 2018.

    Tutorial for a complete yet simple TensorFlow application that showcases the use of APIs necessary to build industry-grade Machine Learning solutions. Also showcases the use of Sonnet library for model building.

    Poništi
  17. 12. ruj 2018.

    More than a positive step towards the democratization of ML, what-if is a boon for AI researchers whose primary objective is to ask what-if...

    Poništi
  18. proslijedio/la je Tweet
    11. ruj 2018.

    What is innate and what is acquired in social cognition? Great discussion of what "innate" means and does not mean. by Video:

    Poništi
  19. proslijedio/la je Tweet

    Excited to announce the release of another TFX component: Data Validation -- an open-source library that helps developers understand, validate, and monitor their ML data at scale. Learn more ↓

    Poništi
  20. proslijedio/la je Tweet
    5. ruj 2018.

    Announcing the launch of Dataset Search, a new way for researchers to find the datasets they need, wherever they’re hosted, whether it’s a publisher's site, a digital library, or an author's personal web page. Learn more at

    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:

    ·