AlphaGo

Поделиться
HTML-код
  • Опубликовано: 25 июн 2024
  • This video explains the details behind AlphaGo! AlphaGo uses policy and value networks to reduce the search space in MCTS! Thanks for watching! Please Subscribe!
    Paper Link: www.nature.com/articles/natur...
  • НаукаНаука

Комментарии • 15

  • @connorshorten6311
    @connorshorten6311  4 года назад +3

    0:54 Quick Overview
    3:55 Tree Search Complexity
    4:44 Training Policy Networks with Supervised Learning
    5:44 State Representation
    6:18 Policy Networks with Self-Play and Policy Gradients
    7:33 Training Value Network
    8:25 Monte Carlo Tree Search
    12:26 Results of AlphaGo
    12:52 Distributed Implementation

  • @BiancaAguglia
    @BiancaAguglia 4 года назад +2

    The other day I watched Francois Chollet's interview on Lex Fridman's YT channel, and saw your comment about how Keras was one of the reasons you didn't give up on studying DL. I'm glad you stuck with it: you do a wonderful job teaching what you're learning. I can't say that I understand everything you're talking about, but I'm getting better every day. Like you, I'm not giving up on learning AI/DL.ML even if it feels overwhelming at times. Thank you for making my learning easier. 😊

    • @connorshorten6311
      @connorshorten6311  4 года назад

      Awesome! I really enjoyed Lex's interview with Francois Chollet, Keras is an amazing API for Deep Learning. It's also been really cool to see the tighter integration between Keras and Tensorflow with later releases, you have access to a lot of really useful / advanced things in Keras like Distributed Training, Mixed Precision, and Quantization. Goodluck with your learning journey!

  • @konstantin6482
    @konstantin6482 3 года назад +1

    Jeeez this is SO good. Dense information flow and clear communication! A man certainly knows what he's talking about

    • @connorshorten6311
      @connorshorten6311  3 года назад

      Thank you so much, I really appreciate it, glad you liked the video!!

  • @josephedappully1482
    @josephedappully1482 4 года назад +1

    Your videos are underrated! Thanks for them, and keep 'em coming! :)

  • @HeduAI
    @HeduAI 11 месяцев назад

    This was a pretty good overview! Thanks ...

  • @FastFSharp
    @FastFSharp 2 года назад

    This is outrageously good!

  • @jessewu4528
    @jessewu4528 3 года назад +2

    I am an undergraduate student, currently taking an intro to AI course
    Our teacher told us to write a project on something we are interested, I choose to write AlphaGo
    Was trying to understand how all those things work, found this video.
    After watching this I kinda get the idea of how it works, but because of lack of knowledge, still a little bit confused.
    Overall, it is a great video, I hope I can learn more by reading paper.

  • @Ree1981
    @Ree1981 4 года назад +9

    Please talk slower. I mean, it's a commonly known thing to teachers and speakers. If you speak too fast, especially in an information heavy subject, a lot just isn't going to stick. A pause of 1 second, no matter how artificial it seems, here and there should help.

    • @connorshorten6311
      @connorshorten6311  4 года назад

      Thank you for the feedback! I will try to be more mindful of this going forward!

    • @metalman1884
      @metalman1884 2 года назад +1

      In case some people don't know, you can adjust your play speed to be slower or faster in any RUclips video.

  • @khalmaxroboticsgh3176
    @khalmaxroboticsgh3176 Год назад

    you speak really fast