Translating lost languages using machine learning
The First Art Newspaper on the Net    Established in 1996 Saturday, December 7, 2024


Translating lost languages using machine learning
In developing a system to help decipher lost languages, MIT researchers studied the language of Ugaritic, which is related to Hebrew and has previously been analyzed and deciphered by linguists. Image courtesy: S.R.K. Branavan.



CAMBRIDGE, MASS.- Recent research suggests that most languages that have ever existed are no longer spoken. Dozens of these dead languages are also considered to be lost, or “undeciphered” — that is, we don’t know enough about their grammar, vocabulary, or syntax to be able to actually understand their texts.

Lost languages are more than a mere academic curiosity; without them, we miss an entire body of knowledge about the people who spoke them. Unfortunately, most of them have such minimal records that scientists can’t decipher them by using machine-translation algorithms like Google Translate. Some don’t have a well-researched “relative” language to be compared to, and often lack traditional dividers like white space and punctuation. (To illustrate, imaginetryingtodecipheraforeignlanguagewrittenlikethis.)

However, researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) recently made a major development in this area: a new system that has been shown to be able to automatically decipher a lost language, without needing advanced knowledge of its relation to other languages. They also showed that their system can itself determine relationships between languages, and they used it to corroborate recent scholarship suggesting that the language of Iberian is not actually related to Basque.

The team’s ultimate goal is for the system to be able to decipher lost languages that have eluded linguists for decades, using just a few thousand words.

Spearheaded by MIT Professor Regina Barzilay, the system relies on several principles grounded in insights from historical linguistics, such as the fact that languages generally only evolve in certain predictable ways. For instance, while a given language rarely adds or deletes an entire sound, certain sound substitutions are likely to occur. A word with a “p” in the parent language may change into a “b” in the descendant language, but changing to a “k” is less likely due to the significant pronunciation gap.

By incorporating these and other linguistic constraints, Barzilay and MIT PhD student Jiaming Luo developed a decipherment algorithm that can handle the vast space of possible transformations and the scarcity of a guiding signal in the input. The algorithm learns to embed language sounds into a multidimensional space where differences in pronunciation are reflected in the distance between corresponding vectors. This design enables them to capture pertinent patterns of language change and express them as computational constraints. The resulting model can segment words in an ancient language and map them to counterparts in a related language.




The project builds on a paper Barzilay and Luo wrote last year that deciphered the dead languages of Ugaritic and Linear B, the latter of which had previously taken decades for humans to decode. However, a key difference with that project was that the team knew that these languages were related to early forms of Hebrew and Greek, respectively.

With the new system, the relationship between languages is inferred by the algorithm. This question is one of the biggest challenges in decipherment. In the case of Linear B, it took several decades to discover the correct known descendant. For Iberian, the scholars still cannot agree on the related language: Some argue for Basque, while others refute this hypothesis and claim that Iberian doesn’t relate to any known language.

The proposed algorithm can assess the proximity between two languages; in fact, when tested on known languages, it can even accurately identify language families. The team applied their algorithm to Iberian considering Basque, as well as less-likely candidates from Romance, Germanic, Turkic, and Uralic families. While Basque and Latin were closer to Iberian than other languages, they were still too different to be considered related.

In future work, the team hopes to expand their work beyond the act of connecting texts to related words in a known language — an approach referred to as “cognate-based decipherment.” This paradigm assumes that such a known language exists, but the example of Iberian shows that this is not always the case. The team’s new approach would involve identifying semantic meaning of the words, even if they don’t know how to read them.

“For instance, we may identify all the references to people or locations in the document which can then be further investigated in light of the known historical evidence,” says Barzilay. “These methods of ‘entity recognition’ are commonly used in various text processing applications today and are highly accurate, but the key research question is whether the task is feasible without any training data in the ancient language.”

The project was supported, in part, by the Intelligence Advanced Research Projects Activity (IARPA).

Reprinted with permission of MIT News










Today's News

October 25, 2020

Art auctions embrace a future of socially distant bidding

Sotheby's to present largest private collection of Ansel Adams photographs this December

Artsy announces partnership with Artlogic to benefit both organizations' gallery networks

Cindy Sherman presents ten new photographs at Metro Pictures

Fondation Cartier pour l'art contemporain presents an immersive installation created by Sarah Sze

Putting pencil to paper, in galleries and in the voting booth

Pandemic-forced isolation opens new artistic pathways

British Museum welcomes the Iraqi Prime Minister Mustafa Al-Kadhimi and a 100-year research project concludes

Hindman's Sports Memorabilia auction sells 91% of lots offered

Iconic UK arts institutions get £ 75m virus funding

Reimagining Lady Liberty's torch to meet this moment

High Museum opens first-ever comprehensive survey of Julie Mehretu's career

Scientific pioneers, including Louis Pasteur, make their mark on Heritage Auctions' manuscripts event

Dawit L. Petros now represented by Bradley Ertaskiran

Petersen Automotive Museum named 2020 Museum Of The Year

Exceptional works by Richter and Calder lead Bonhams Post War & Contemporary Art sale in New York

Major works added to Kemper Museum of Contemporary Art's permanent collection

von ammon co opens an exhibition of works by Timur Si-Qin

Translating lost languages using machine learning

Success for Frank Auerbach at Bonhams Post-War & Contemporary Art sale in London

Steidl publishes 'Harmony Korine, Juergen Teller: William Eggleston 414'

Sotheby's Wine announces first offering of Japanese Sake

Exhibition at Smack Mellon celebrates the 100th anniversary of the ratification of the 19th Amendment

Want To Experience The Thrill Of Gambling Online? Play Online Poker!!!

How To Get Your pH Balance Back On Track

How Pet Fit app is beneficial for pet owners?

Going against convention with abstract art

Do You Pay Child Support with Joint Custody [Explained]

How Does Custody Work With Breastfeeding: Is It Even Possible?

How does meat get heated in a frying pan

How to determine the value of a personal injury case? (Step by Step)

How to Use the ICBC Insurance Calculator




Museums, Exhibits, Artists, Milestones, Digital Art, Architecture, Photography,
Photographers, Special Photos, Special Reports, Featured Stories, Auctions, Art Fairs,
Anecdotes, Art Quiz, Education, Mythology, 3D Images, Last Week, .

 



Founder:
Ignacio Villarreal
(1941 - 2019)
Editor & Publisher: Jose Villarreal
Art Director: Juan José Sepúlveda Ramírez
Writer: Ofelia Zurbia Betancourt

Attorneys
Truck Accident Attorneys
Accident Attorneys
Houston Dentist
Abogado de accidentes
สล็อต
สล็อตเว็บตรง
Motorcycle Accident Lawyer

Royalville Communications, Inc
produces:

ignaciovillarreal.org juncodelavega.com facundocabral-elfinal.org
Founder's Site. Hommage
to a Mexican poet.
Hommage
       

The First Art Newspaper on the Net. The Best Versions Of Ave Maria Song Junco de la Vega Site Ignacio Villarreal Site Parroquia Natividad del Señor
Tell a Friend
Dear User, please complete the form below in order to recommend the Artdaily newsletter to someone you know.
Please complete all fields marked *.
Sending Mail
Sending Successful