Getting started with open source game playing AIs

on YouTube ↗28m

" summary

Stockfish is an open source C++ engine ranked at about 3.3k Elo and wins by running thousands of self-play games through a distributed testing network to measure tiny heuristic tweaks. Go engines like Pachi use Monte Carlo tree search plus neural nets to sample random playouts instead of hard heuristics. A Java GPLv3 project implements Magic the Gathering with a UI and handles hidden hands by sampling hidden-information (randomly instantiating unknown cards) so simulations can run and the AI makes decisions.

AI-written summary · about

" chapters (8)

1. 0:00Introduction to open source game AIs
2. 1:01History of computer game playing
3. 3:27Stockfish and open source chess
4. 8:44Minimax algorithm explained
5. 14:39Heuristics and testing in game AIs
6. 16:26AlphaGo and Monte Carlo tree search
7. 22:45Hidden information in card games
8. 26:56Summary and open source projects

" related talks

• 15mAI Self - Learning Game PlayingRichard Abbuhl · Devoxx Belgium 2017
• 46mBeating Go thanks to the power of randomnessTobias Pfeiffer · RubyConf 2015
• 18mJavaScript and Artificial IntelligenceAngel "Java" Lopez · JSConf.ar 2014
• 46mAI TakeoverNicolas Greffard, Tanguy Gasselin · DevFest Nantes 2018
• 18mBuilding a Chess CoachAnant Dole, Asbjorn Steinskog · AI Engineer Europe 2026
• 36mPlaying Go with ClojureZach Tellman · Clojure/conj 2012
• 56mCan I do AI?Izzet Mustafaiev · JavaDay Ukraine 2017
• 15mChess in the Age of Artificial IntelligenceLaurent Fressinet · Devoxx France 2026