Recreating DeepMind's AlphaZero - AI Plays Connect 4 - Part 5: Building the ResNet Neural Network

Поділитися
Вставка
  • Опубліковано 27 лип 2024
  • In this series we re-create DeepMind's AlphaZero to create an Artificial Intelligence Reinforcement Learning algorithm that plays Connect 4 using Python.
    View the rest of the series: ua-cam.com/users/playlist?list.
    Full Code: github.com/ThePrincipalCompon...
    AI, Deep Learning, Reinforcement Learning, DeepMind, Convolutional Neural Networks, Monte Carlo Tree Search, AlphaGo, AlphaZero, Connect4, ConnectX
  • Наука та технологія

КОМЕНТАРІ • 16

  • @gruhitpatel9061
    @gruhitpatel9061 4 місяці тому

    Great Series and excellent explanation. Please release part-6 soon.

  • @theinvisibleman9582
    @theinvisibleman9582 Рік тому +4

    Hey there @ThePrincipalComponent . This was an awesome series. Can you possibly complete it soon? I was taking inspiration from this series to make another project for my school on a similar theme and now I am stuck. It would be awesome if you release the next part soon, I believe it is almost done.

  • @askent2250
    @askent2250 Рік тому +2

    I really like your videos, looking forward to part 6

  • @factoral2645
    @factoral2645 Рік тому +1

    Great series! Looking forward to the next part

  • @juanolivares-ruiz8028
    @juanolivares-ruiz8028 Рік тому +5

    Great series! When will part 6 be released?

  • @kimjong-un4521
    @kimjong-un4521 Місяць тому

    Part 6 please

  • @jjjay4559
    @jjjay4559 Рік тому

    Cant wait to see part 6

  • @craiglobo2165
    @craiglobo2165 Рік тому +1

    I really hope you release part 6

  • @adolfjamesurian6510
    @adolfjamesurian6510 Рік тому

    Are you going to use self-play algoritm to train the network with the help of mcts? If ever you have plan in making part 6.

  • @xelastone
    @xelastone 4 місяці тому

    Will you be completing the series?

  • @baderalhamiri4313
    @baderalhamiri4313 Рік тому +3

    Hello, Will you be releasing the 6th part anytime soon?

  • @bjornnorenjobb
    @bjornnorenjobb Рік тому

    A question, is this the same architecture of the CNN/ResNet as the implementation that DeepMind implemented?

    • @theprincipalcomponent
      @theprincipalcomponent  Рік тому +1

      The architecture in the video is a much smaller version of it, shapes are also much different since Go is a larger more complicated game. But this could be easily scaled to their implementation

    • @bjornnorenjobb
      @bjornnorenjobb Рік тому

      @@theprincipalcomponent Okey, thank you! I love your content and think it is awesome. 🦾

  • @albertvanbreemen7081
    @albertvanbreemen7081 Рік тому

    Hi! Is there a final vid to this series? Sid it learn to play connect4?

  • @danidiaz9510
    @danidiaz9510 Рік тому

    Oh no, where's the next part?