MarI/O - Machine Learning for Video Games
What it is
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.
Why Hacker News recommends it
The go-to accessible introduction to neuroevolution and NEAT, praised for crediting prior research.
- 10 people recommended it
- 10 different months
- 3 strong recommendations
- first 2015, last 2026
What HN commenters said
“Seth's [Show More] Links on the MarI/O video[1] are very very comprehensive and (IMHO) indeed a great example of scholarship !”
Video explains NEAT and credits prior research thoroughly.
deepnet, 2018-04 · Strong recommendation · in “The differences between tinkering and research (2016)”
“There also SethBling's excellent video on YouTube about machine learning specifically with Super Mario World”
Excellent video on machine learning with Super Mario World.
dberst, 2024-09 · Strong recommendation · in “Synchronizing Pong to music with constrained optimization”
“Here is my favorite introduction to NEAT”
Favorite introduction to NEAT.
andai, 2025-06 · Strong recommendation · in “Self-Adapting Language Models”
“his brief videos give a really nice introduction to the topic that should be accessible to most people”
Brief accessible introduction to neuroevolution playing Mario.
pdkl95, 2015-06 · Recommendation · in “Hacking Super Mario World to Warp to the Credits [video]”
“There's a great video that made the rounds last year about building a neural net that plays Super Mario World”
Helps visualize how a neural net learns to play Mario.
slpsys, 2016-03 · Recommendation · in “Ask HN: What are some examples of beautiful software?”
“(great educational video)”
Educational video on ML playing Mario.
notgood, 2016-11 · Recommendation · in “NeuroEvolution – Flappy Bird”
“This video of a Super Mario playing AI was pretty neat”
Neat demo of an AI learning Super Mario.
new299, 2016-12 · Recommendation · in “Whence your reward function?”
“after watching a Youtube video of an implementation of it”
Inspired commenter to implement NEAT as a teenager.
intunderflow, 2022-04 · Recommendation · in “Ask HN: Which algorithms based on biological ideas do you know?”
Excerpts from Hacker News comments, linked to the originals. Summaries and labels were written by a language model from those comments.
Topics: neat, genetic algorithms, neural networks, super mario world, machine learning
Also recommended by the same people
explainerDeadly Truth of General AI? - ComputerphileComputerphile · 3 HN recommenders · 2016–2024
explainerGender Neutral Pronouns: They're Here, Get Used To ThemTom Scott · 3 HN recommenders · 2013–2015
talkBlueHat IL 2019 - Andrew "bunnie" Huang - Supply Chain Security: "If I were a Nation State...”Microsoft Israel R&D Center · 6 HN recommenders · 2019–2024
explainerBut what is the Fourier Transform? A visual introduction.3Blue1Brown · 16 HN recommenders · 2018–2025
talkCybersecurity as Realpolitik by Dan Geer presented at Black Hat USA 2014Black Hat · 1 HN recommender · 2014–2023
lectureMark Blyth ─ Global TrumpismWatson School of International and Public Affairs · 3 HN recommenders · 2016–2019
documentaryChernobyl 3828TeleconStudio · 2 HN recommenders · 2014–2019
commentaryFlat Earth: A Measured Responsehbomberguy · 3 HN recommenders · 2019–2024