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
-
November 15, 2024
![a hand with robotic elements attached is shown.]()
National conference celebrates interdisciplinary collaborations and research
More than 170 higher education professionals from across the country came to RIT Nov. 14-16 for the 2024 a2ru National Conference. The a2ru national conference is an opportunity for practitioners and researchers from across higher education to share innovations and perspectives in the arts.
-
November 11, 2024
![Arthur Deane walks through a park like setting in front of the capitol building wearing a dark polo shirt and khaki pants.]()
Amazon executive supports new path for cybersecurity students
Arthur Deane broke into the cybersecurity industry and he hasn’t looked back. Now, he’s helping others unlock opportunities in cybersecurity, paving the way for a more diverse and inclusive tech industry
-
November 8, 2024
![An illustration of a brain with the words Science Advances displayed over the top in white, all on a dark blue background.]()
RIT professor proposes new way to make artificial intelligence smarter and greener
The brain is a great source of inspiration for Alexander Ororbia, an assistant professor of computer science and cognitive science at RIT. By mimicking how neurons in the brain learn, Ororbia is working to make artificial intelligence more powerful and energy efficient. His research was recently published in the journal Science Advances.
-
November 7, 2024
![a man in a red fedora and plaid shirt sits on a curb in a city setting.]()
Open source opens doors at Red Hat
As a Fedora community architect, Justin Flory wears many hats. He is a community liaison, accountant, and project manager—all rolled into one.
