Rezultati pretraživanja
  1. 28. sij

    Is Meta-Learning the New Black? by ——— ——— See my article "Meta-Learning for Better Machine Learning":

  2. 28. sij
  3. 17. pro 2018.

    Want to know what likelihood-free inference is and how can help science? Read our new article w/ : distilling our recent paper Recurrent Machines for Likelihood-Free Inference published at the NeurIPS 2018 MetaLearn workshop

  4. 18. stu 2019.
  5. 17. lip 2019.

    has more interest than anticipated.

  6. 5. ruj 2019.

    Check out our recent work on the convergence theory of Model-Agnostic (MAML) methods on arXiv: Joint work with and Asu Ozdaglar.

    Prikaži ovu nit
  7. 19. stu 2019.

    I learned it takes a good skill to simplify complex topics. Teaching or training someone is hard haha xD But I also realized what I know, what I don't and can still improve by teaching.

    Prikaži ovu nit
  8. 31. sij
  9. 30. sij

    Succeed by adding empowering techniques to the memory that boost its processing ! A lot of advances in memory research suggest methods that can be applied to enhance educational practices. Read more

  10. 28. sij
  11. 19. sij
  12. may be the new black, but what are the 3 critical components machines need in order to master ? The Dean of explains :

  13. Yoshua Bengio in keynote describes his vision for moving from system 1 to system 2 intelligence. Lots of trendy topics , , , etc. + a reading rec for Daniel Kahnemann's Thinking Fast and Slow.

  14. 3. pro 2019.

    So many cool new benchmarking environments. This will also surely foster research! Procgen (games): Metadataset (image classification): MetaWorld (robotic manipulation): The future is bright

  15. 4. stu 2019.
  16. Zero-Shot Task Adaptation by Homoiconic Meta-Mapping cc

  17. 25. ruj 2019.

    Integrating and and other higher order skills through bridging the gaps between learning

  18. 14. ruj 2019.

    Last call :). If you are looking for a post-doctoral fellowship to work on and (), please let me know right away!

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.