Alexander Ororbia to Work in Collaboration with Google Research
Abstract/Mission:
Backprop-Free Training for Models in Natural Language Processing
For this project Alexander Ororbia will work in collaboration with Google Research will focus on developing a backprop-free method for training neural language architectures. Specifically, he is interested in adapting some of his more recently proposed procedures for representation learning in natural language processing (NLP), with a focus on unsupervised and semi-supervised learning (of useful distributed representations of constructs such as documents). Given the expense involved in creating labeled data, especially when working with text data, an effective alternative to backprop would be of interest to Google in terms of annotation cost reduction and, furthermore, to exploit the underlying parallelism when large-scale computational resources are available, i.e., massive clusters of GPUs, CPUs, and/or TPUs.
Recommended News
-
December 13, 2024
![a college age girl in a black floral dress sits at a table in a lab room towards the right side of the image]()
Student spotlight: Fulbright student's research applies AI to real-world problems
Diana Velychko, a Fulbright master’s student in artificial intelligence, is paving the way for more intuitive artificial intelligence (AI) systems. Her work focuses on the intersection of human perception and AI, aiming to bridge the gap between how people see the world and how machines interpret it.
-
December 12, 2024
![people sit on both sides of a fashion show runway where a person in a dress in clown styling walks down the center.]()
Designers leave nothing on the table during the Beyond Fashion showcase
Decked out with technicolor lights and vibrant animated displays, RIT’s MAGIC Spell Studios soundstage became a runway last week during Beyond Fashion. The rainbow runway featured a diverse collection of garments and wearables crafted by local designers and RIT students across multiple disciplines.
-
December 10, 2024
![Jonathan Weissman appears on TV with a chryon stating his name below him.]()
Malicious email, purportedly from PayPal; slips through filter; Pennsylvania woman bilked out of $15,000
WHTM-TV talks with Jonathan Weissman, principal lecturer in the Department of Cybersecurity, about the scams.
-
December 9, 2024
![Matthew Wright stands in front of a camera on TV with his name appearing below his face.]()
Deepfakes: Why we’re seeing more of them and how to differentiate reality from fiction
WROC-TV speaks to Matthew Wright, endowed professor and department chair in the Golisano College of Computing and Information Sciences, and Computing and information sciences Ph.D. student Y. Kelly Wu about identifying deepfakes.
