The Sumerian language spawned the earliest form of writing around 5000 years ago in Mesopotamia (modern-day Iraq). However, time has taken its toll on these early writings, inscriptions, and instructions. Historians have long been concerned about the disappearance of manuscripts that could provide insight into ancient civilisation’s life and culture; Artificial Intelligence has now come to their rescue.
The first deep neural network, named after Ithaca, a Greek island in Homer’s Odyssey, will aid in not only reconstructing the lost text of damaged inscriptions, but also identifying their original location and determining the date they were inscribed. This AI, which was created to aid and broaden the historian’s workflow, has reached a 62% accuracy rate while repairing damaged texts and has enhanced historians’ accuracy from 25% to 72%.
In a study published in the journal Nature, researchers said that models such as Ithaca can unlock the cooperative potential between artificial intelligence and historians, transformationally impacting the way that we study and write about one of the most important periods in human history.
Deep neural networks, which are based on biological neural networks, can find and exploit complicated statistical patterns in large amounts of data. Ithaca is an example of a project that combines technology, supercomputing, and ancient history to disclose previously unknown truths concealed in plain sight.
Ithaca was taught to conduct textual restoration, geographical attribution, and chronological attribution all at the same time. Between the seventh century BC and the fifth century AD, researchers trained the algorithm on inscriptions written in the ancient Greek language and across the ancient Mediterranean globe.
The architecture of Ithaca was carefully tailored to each of the three epigraphic tasks, meaningfully handling long-term context information and producing interpretable outputs to enhance the potential for human-machine cooperation. “We believe machine learning could support historians to expand and deepen our understanding of ancient history, just as microscopes and telescopes have extended the realm of science Yannis Assael, Staff Research Scientist at DeepMind said in a statement.
Historians have already utilised Ithaca to clarify current historical debates, such as the dating of a series of significant Athenian decrees assumed to have been written before 446/445 BCE. The average anticipated date for the decrees in Ithaca is 421 BCE, which matches the new evidence and shows how machine learning may help with historical conflicts.
“Although it might seem like a small difference, this date shift has significant implications for our understanding of the political history of Classical Athens. We hope that models like Ithaca can unlock the cooperative potential between AI and the humanities, transformationally impacting the way we study and write about some of the most significant periods in human history,” Thea Sommerschield, Marie Curie Fellow at Ca’ Foscari University of Venice and fellow at Harvard University’s CHS said.
Historians are now working on other versions of the AI, which has been trained in different ancient languages to study other ancient writing systems, from Akkadian to Demotic and Hebrew to Mayan.
