У нас вы можете посмотреть бесплатно A.I. Learns to Beat QWOP (World's HARDEST Game) или скачать в максимальном доступном качестве, видео которое было загружено на ютуб. Для загрузки выберите вариант из формы ниже:
Если кнопки скачивания не
загрузились
НАЖМИТЕ ЗДЕСЬ или обновите страницу
Если возникают проблемы со скачиванием видео, пожалуйста напишите в поддержку по адресу внизу
страницы.
Спасибо за использование сервиса ClipSaver.ru
I taught an AI how to beat QWOP! In fact it was so good, it also got onto the speed run leaderboard. Inspired by popular videos on teaching AI how to walk like that by Code Bullet and Dani, I decided to apply machine learning techniques to teach an AI to beat QWOP. This was a rollercoaster ride with techniques such as evolutionary algorithms, NEAT, and imitation learning. Enjoy! Code - https://github.com/CoderboiYT/QWOPAI Original QWOP game - http://www.foddy.net/Athletics.html Genetic algorithms in python tutorial series - • Neural nets with Flappy and pygame Check out this legend that actually got an AI to beat the speed run record! I also got a lot of help from his code so give him some likes and stars - https://github.com/Wesleyliao/QWOP-RL Chapters: 00:00 - 01:23 Introduction 01:23 - 02:25 Experimenting with evolution 02:25 - 04:11 Getting to the finish line using NEAT 04:11 - 05:38 Getting a speedrun using imitation learning 05:38 - 08:18 Final training footage 08:18 - 10:41 Final result 10:41 - 11:14 Conclusion