Prof. Ion Androutsopoulos and researchers John Pavlopoulos and Marita Chatzipanagiotou, contributed to a significant work published in Nature journal

    Fig. 1

    "This inscription (Inscriptiones Graecae, volume 1, edition 3, document 4, face B听(IG听I3听4B)) records a decree concerning the Acropolis of Athens and dates to 485/4听BC. Marsyas, Epigraphic Museum, WikiMedia CC BY 2.5."

    听

    Predicting the past with Ithaca
    听

    Restoring, placing, and dating ancient texts through collaboration between AI and historians

    听

    Prof. Ion Androutsopoulos and researcher John Pavlopoulos, from AUEB's , along with Marita Chatzipanagiotou, student (now graduate) of AUEB's , are co-authors of the paper published on听9/3/2022 in :

    Predicting the past with Ithaca
    Restoring, placing, and dating ancient texts through collaboration between AI and historians

    Yannis Assael [1], Thea Sommerschield [2,3]
    Brendan Shillingford [1], Mahyar Bordbar [1], John Pavlopoulos [4], Marita Chatzipanagiotou [4], 听Ion Androutsopoulos [4], Jonathan Prag [5], Nando de Freitas [1]

    [1] DeepMind, United Kingdom; [2] Ca鈥� Foscari University of Venice, Italy; [3] Center for Hellenic Studies, Harvard University, USA; [4] 黑料科 of Economics and Business, Greece; [5] University of Oxford, United Kingdom

    "...In line with听听of solving intelligence to advance science and humanity, we collaborated with the听, the听, and the听听to explore how machine learning can help historians better interpret these inscriptions 鈥� giving a richer understanding of ancient history and unlocking the potential for cooperation between AI and historians.

    In a听听published today in听Nature, we jointly introduce 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. Ithaca is named after the Greek island in听听and builds upon and extends听, our previous system that focused on textual restoration. Our evaluations show that Ithaca achieves 62% accuracy in restoring damaged texts, 71% accuracy in identifying their original location, and can date texts to within 30 years of their ground-truth date ranges. Historians have already used the tool to reevaluate significant periods in Greek history..."

    Read the paper and find more information:

    Paper:听
    Video:听
    Blog:听
    Blog in Greek:听听

    听听

    听

    ID: 
    3364

    Last updated: 12 March 2022