I am a computer science PhD student at Georgia Tech, and am advised by Prof. Humphrey Shi. I'm primarily interested in attention-based models, focusing on their application in computer vision. It spans all the way from infrastructure and system co-design, to designing new architectures based on attention for computer vision. My longest running project has been providing fast and efficient implementations of Neighborhood Attention, a sliding window attention primitive that allows fine-grained control over attention span, drawing a spectrum of possible attention patterns between linear projection and self attention. Check out NATTEN for more details.
GitHub / Google Scholar / Twitter / LinkedIn
Select publications & preprints
2024
Faster Neighborhood Attention: Reducing the O(n^2) Cost of Self Attention at the Threadblock Level
Ali Hassani, Wen-Mei Hwu, and Humphrey Shi.
2023
Neighborhood Attention Transformer
Ali Hassani, Steven Walton, Jiachen Li, Shen Li, and Humphrey Shi.
CVPR 2023.
2022
StyleNAT: Giving Each Head a New Perspective
Steven Walton, Ali Hassani, Xingqian Xu, Zhangyang Wang, and Humphrey Shi.
2021
Experience
06/2023 - 12/2023
Software Engineering Intern
06 - 09/2022
Picsart AI Research
Research Intern
Since 03/2021
Graduate Researcher
2019 to 2021
Mahani Mathematical Research Center
Undergraduate Researcher
Teaching
Education
2016-2020
B.S. in Computer Science
University of Kerman