DeepMind’s Ithaca model is named after the Greek island in Homer’s Odyssey.

DeepMind’s new Ithaca AI model will help historians in restoring, placing, and dating ancient Greek texts, marking an era of productive collaboration between AI and researchers of the magnificent Greek civilization. (DeepMind)

There is a wealth of historical texts from 2,500 years ago, when the Greeks took to inscribing on stone, pottery and metal. Their writings spanned laws, calendars, leases, even oracles. With the passage of so much time, however, these ancient texts, wherever available, are often damaged or found at places far removed from their origin.

Historians are, therefore, hampered in their quest for understanding of the Greek civilization because the record is either incomplete or defaced.

Ithaca from DeepMind

DeepMind collaborated with the Department of Humanities of Ca’ Foscari University of Venice, the Classics Faculty of the University of Oxford, and the Department of Informatics of the Athens University of Economics and Business to solve this problem using artificial intelligence.

The machine learning tool was trained on the largest digital dataset of Greek inscriptions from the Packard Humanities Institute.

The effort builds further upon DeepMind’s Pythia, an earlier system for textual restoration.

The result: Ithaca, the first deep neural network that can restore the missing text of damaged inscriptions, identify their original location, and help establish the date they were created.

Here are ‘before’ and ‘after’ images sourced from the DeepMind blog:

 On testing, Ithaca delivered an accuracy of 62% in text restoration, 71% in pinpointing the origin, and dating to within 30 years.

“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,” writes DeepMind.

Related Story: Now AI Could Unlock Secrets From 2,500-Year Old Iranian Tablets