MarI/O - Machine Learning for Video Games
10 人推荐 · 跨 10 个月 · 2015–2026
中文导读
知名博主 SethBling 演示自己开发的 MarI/O 程序：利用神经网络与 NEAT 遗传算法，让程序从零自主学会通关《超级马力欧世界》。视频不仅展示了直观的演化过程，还在简介中详尽附上了源码、论文与学术背景。
为什么 HN 推它
关于神经演化与 NEAT 算法最通俗直观的现象级入门作，尊重前人学术脉络的严谨科普态度同样广受好评。
「这期精炼短小的小短片不仅是顶级的科普入门，而且对绝大多数门外汉来说也毫无理解门槛。」HN 网友原话：his brief videos give a really nice introduction to the topic that should be accessible to most people
英文原摘要（HackerTube / Claude 整理）
SethBling shows MarI/O, a program using neural networks and the NEAT genetic algorithm to learn to play Super Mario World. The description links the source code, the paper and background reading.
The go-to accessible introduction to neuroevolution and NEAT, praised for crediting prior research.
HN 上的提及
his brief videos give a really nice introduction to the topic that should be accessible to most people
Example of neural net demos that appear intelligent.
There's a great video that made the rounds last year about building a neural net that plays Super Mario World
Example of ML research on a game.
(great educational video)