logo
ResearchBunny Logo
Abstract
This paper introduces Ithaca, a deep neural network designed to restore damaged ancient Greek inscriptions, attribute them geographically, and chronologically. Ithaca's architecture prioritizes collaboration with historians, decision support, and interpretability. When used by historians, Ithaca improved their accuracy in text restoration from 25% to 72%, achieving 62% accuracy independently. It also demonstrated 71% accuracy in geographical attribution and dated inscriptions to within 30 years of their actual dates, impacting key historical debates.
Publisher
Nature
Published On
Mar 10, 2022
Authors
Yannis Assael, Thea Sommerschield, Brendan Shillingford, Mahyar Bordbar, John Pavlopoulos, Marita Chatzipanagiotou, Ion Androutsopoulos, Jonathan Prag, Nando de Freitas
Tags
ancient Greek inscriptions
deep neural network
text restoration
geographical attribution
historical accuracy
interpretability
historical debates

Related Publications

Explore these studies to deepen your understanding of the subject.

Bioplastic design using multitask deep neural networks
Environmental Studies and Forestry

Bioplastic design using multitask deep neural networks

C. Kuenneth, J. Lalonde, et al.

Toward automated classification of monolayer versus few-layer nanomaterials using texture analysis and neural networks
Physics

Toward automated classification of monolayer versus few-layer nanomaterials using texture analysis and neural networks

S. H. Aleithan and D. Mahmoud-ghoneim

Internal pipe corrosion assessment method in water distribution system using ultrasound and convolutional neural networks
Engineering and Technology

Internal pipe corrosion assessment method in water distribution system using ultrasound and convolutional neural networks

Y. Sung, H. Jeon, et al.

Design and Analysis of a Deep Learning Ensemble Framework Model for the Detection of COVID-19 and Pneumonia Using Large-Scale CT Scan and X-ray Image Datasets
Medicine and Health

Design and Analysis of a Deep Learning Ensemble Framework Model for the Detection of COVID-19 and Pneumonia Using Large-Scale CT Scan and X-ray Image Datasets

X. Xue, S. Chinnaperumal, et al.

Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs—just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny