US Edition
AI Reconstructs Rules for Ancient Roman Board Games
Thousands of AI simulations indicate the carved limestone was likely a Roman blocking game named Ludus Coriovalli, predating known European examples by centuries.
- In a paper in Antiquity published Monday, researchers report AI simulations suggest the Heerlen stone was a Roman blocking game, using the Ludii AI system.
- A 212 by 145 millimetre limestone slab was found in Coriovallum, now Heerlen, dating to AD 250–476; the carved rectangle with four diagonal and one straight line shows wear from sliding playing pieces.
- Using Ludii, two AI agents replayed thousands of possible games testing 130 rule variations and over 100 rule sets, producing nine blocking-game rule sets consistent with the wear.
- The team tentatively calls the reconstruction Ludus Coriovalli and published a playable online version, which if confirmed would push evidence for blocking games in Europe back several centuries.
- Some experts remain unconvinced that the object can be proven a game board, as the board’s archaeological context is unknown and certainty may be impossible, but researchers advocate combining AI and archaeology methods.
62 Articles
62 Articles
Stone is Roman gameboard and AI has worked out the rules - DutchNews.nl
Stone is Roman gameboard and AI has worked out the rules
The stone is strikingly flat, decorated with lines and probably once served as a playing board: But for which game exactly remained unclear for a long time – until a Dutch archaeologist let an AI compete.
The aim of the game is to block the other player's pawns.
Scientists Unlock the Rules of Long-Lost Roman Board Game
Ancient Roman board game. Credit: Trey Jones / Wikimedia Commons / CC BY-SA 3.0 Scientists have used artificial intelligence to reconstruct rules of a long-forgotten Roman board game that was likely played, offering new insight into everyday life in the ancient world and pushing the origins of a known game type back by centuries. The discovery centers on a carved limestone object identified in 2020 by archaeologist Walter Crist while examining i…
Coverage Details
Bias Distribution
- 40% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
