[Colloquium] Today: Gil Rosenthal MS Presentation/May 17, 2023

[Jessica Garza]

This is an announcement of Gil Rosenthal's MS Presentation

Gil Rosenthal is a student in the Bx/MS program.

———————————————————————————————————————————

Date: Wednesday, May 17, 2023

Time: 3 PM, CST

Location: JCL 298

Zoom: https://uchicago.zoom.us/j/91727524850?pwd=d3ZTZjdURGpVMXhsWk9BM2FZVWVKQT09 <https://urldefense.com/v3/__https://uchicago.zoom.us/j/91727524850?pwd=d3ZTZjdURGpVMXhsWk9BM2FZVWVKQT09__;!!BpyFHLRN4TMTrA!_VsA5enRfQfwQHgtyYAffUj4Q0u07dTGTa0kqMp17kXz2eK90oSr4GAyuO-BxeKnMJDEf5D3S7CprOuvmrPQY-4BU4HX$>

M.S. Candidate: Gil Rosenthal

M.S. Paper Title: Machina Cognoscens: Neural Machine Translation for Latin, a Case-Marked Free-Order Language

Advisor: Allyson Ettinger

Committee Members: Allyson Ettinger, Jeff Tharsen, and Chenhao Tan

———————————————————————————————————————————

Abstract:

Neural methods have brought a revolution in automated Machine Translation processes, with most highly-spoken languages having robust training datasets and near-human performance. However, these methods have lacked the same effect in Case-Marked Free-Order languages. A free-order language is one that has no specific word order, i.e. the subject, verb, and object can be anywhere in the sentence without violating the rules of the grammar. Case-marked means that additional information about the word, such as the number and function, are encoded in morphological features of the word, such as case or conjugation. As a target language, we use Latin, which is a FOCM language with extremely poor machine translation tools existing. We have created a first-of-its-kind Parallel Translation Dataset consisting of roughly 100k pairs, and evaluated its performance in Neural Machine Translation, with novel methods of preprocessing to encode morphology, and new approaches to transfer learning. We achieve a best performance BLEU of 22.4 on the test dataset, which beats the current State of The Art Google Translate model by over 4.2 BLEU.

———————————————————————————————————————————





Jessica Garza
Assistant Director of Undergraduate Studies
Department of Computer Science
The University of Chicago
John Crerar Library 374
Office: (773) 702-2336

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20230517/96770282/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Gil Rosenthal - MS Paper.pdf
Type: application/pdf
Size: 540636 bytes
Desc: not available
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20230517/96770282/attachment-0001.pdf>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20230517/96770282/attachment-0003.html>