Physical Address

304 North Cardinal St.
Dorchester Center, MA 02124

AI Identifies Author of Charred Scroll Buried by Vesuvius for 2,000 Years


For the first time, researchers identified the author and title of a document locked in a leap in a leap in about 2000 years.

Scroll, Pherc. 172, 79 was restored in the ruins of Herculaneum, the ancient city of Rome, buried in Mount Vesuvius in the Council of Europe. Sliding is now one of the three Herkulaneum Scrolls living in Oxford’s Bodleian Libraries.

Thanks to the study of high resolution scans and some serious intelligent machines, scientists can read almost the papper “Open and read”: In the evilsEpicure by philosopher philodemus.

Contract – has full name Bad and their opposite virtues and about who and what isaccording to Fine Books magazineBasically, the vice fled from the vice and exploring how virtuous life is mainly helped. In the first century, the Philippines wrote work in the first century in the century, and this is read for the first time since the destructive volcano eruption is buried in about 2000 years ago.

The discovery approved by many research groups, project staff, the ancient texts used by the ancient texts of the AI ​​won the first title of $ 60,000 from Vesuvius Challenge.

In recent years, artificial intelligence was a means of decipher the ancient and carbonated landslides, which was first found in the 18th century in the 18th century in the 18th century in 79.

Methods often destroyed them to prevent the fragile, whipped situation, traditional (read: textbook). Now researchers use advanced images and machines that learn to read these texts without opening.

The turning point in 2015, scientists have used an X-ray tomography to scroll an ancient ancient, which creates 3D scanning 3D “unopened”. In this regard, Kentucky University has developed a program that uses a micro-ct image to reveal researchers, volume cardographers, carbon-based compounds, a micro-ct image.

There is no complicated metal, unlike many ancient writing materials, a neural network should have been taught to recognize the fine patterns showing the ink in the carbonized papyrus. Researchers successfully in 2019 displayed This technique sets the stage for extensive applications.

Top down: a reference photo, a texture picture, a forecast image and network generated in the network created.
Top down: a reference photo, a texture picture, a forecast image and network generated in the network created. Photo: Parker et al., Pros One 2019

These progress ended Vesuvius callHe started in 2023, followed the deciphering of unopened scroll. Participants use AI tools – especially using convolutionary neural networks and transformer models – identify and reset the text within the text. In October 2023, the first word (“purple”) was read from an uncovered scroll that won a $ 40,000 award. The call continues, deciphering the addition of additional text and with gifts to improve technology.

Brent seals said that a computer scientist and Vesuvius challenge at Kentucky University Guardian The current swelling of the team is to cleanse, organize and increase the scanning information for researchers who are truly a typical interpretation of the text.

The significant digital drop process is guided by human experience. AI probably emphasizes complex areas in ancient documents, but scientists comment on samples to determine if they form matches or phrases. The goal is to digitize the lost philosophical texts, which are likely to digitize and decipher the ancient texts, both with epicurus or followers.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *