Polaris: Multi-Fidelity Design Space Exploration of Deep Learning Accelerators

C Sakhuja, C Hong, C Lin - arXiv preprint arXiv:2412.15548, 2024 - arxiv.org
This paper presents a tool for automatically exploring the design space of deep learning
accelerators (DLAs). Our main advancement is Starlight, a data-driven performance model …

Incorporating prior knowledge to efficiently design deep learning accelerators

C Sakhuja - 2024 - repositories.lib.utexas.edu
Artificial intelligence (AI) has exploded in popularity over the past decade, and its
computational demand has seen commensurate increase. AI models are getting larger, and …

[PDF][PDF] Use Cases and Deployment of ML in IC Physical Design

A Ghose, AB Kahng, S Kundu, Y Liu, B Pramanik… - 2025 - vlsicad.ucsd.edu
ABSTRACT ML for IC physical design must be deployed in order to have business impacts.
However, deployment in production must navigate many practical considerations, including …

[PDF][PDF] Incorporating Prior Knowledge to Efficiently Design Deep Learning Accelerators

D Chiou, M Erez, A Parashar - cs.utexas.edu
The path to a PhD is mired with taxing twists and turns, and, more than the research, it was
the people around me that kept me going. I owe deep thanks to many, but in this brief …