Omiya Hassan
Boise State University, ID·Engineering
Training and running artificial intelligence (AI) models uses enormous amounts of electricity. Much of that energy goes not into computing but into moving data back and forth between processor chips and memory chips.
- Fields
Engineering
Computer Science
- Topics
AI hardware
artificial intelligence
computer architecture
- Grants
- 1 known grant, through Jul 2029
What they work on
- Co-design an electro-optical application-specific integrated circuit that pairs photonic (light-based) and electronic processing units, assigning AI training and inference to the best-suited unit.
- Develop a heat-aware, high-bandwidth 3D near-memory integration strategy that stacks computing chips vertically close to shared memory to cut data-transfer costs.
- Create a layout-aware data-orchestration and task-scheduling framework, and an open-source architectural simulator, for running AI workloads efficiently across mixed electronic and photonic hardware.
More about the research
This project asks whether that bottleneck can be shrunk by stacking computing chips vertically (3D integration) right next to memory. It also asks whether mixing two kinds of processors, ordinary electronic chips and photonic chips that compute with light, can help, with each type assigned the AI phase it handles best: training or inference (using a trained model).
Why it matters
The energy and infrastructure costs of AI are growing quickly across industry, healthcare, agriculture and finance. Hardware that cuts data-movement costs by placing electronic and photonic processors close to memory could reduce the electricity AI computing needs. An open-source simulator would let researchers and chip designers compare energy and performance trade-offs in electronic-photonic systems.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Qian Yu
University of California-Santa Barbara, CA·Engineering
Develop a highly compressed training method for physics-informed neural networks (PINNs) that uses only forward passes and no backpropagation, so it can run on photonic chips without photonic memory.
Jiaqi Gu
Arizona State University Campus Immersion, AZ·Engineering·Early-career
Develop a learned electromagnetic solver, a machine-learning model of how light moves through photonic devices, and a circuit-level co-simulator for fast simulation of mixed electronic-photonic chips.
Divya Mahajan
Georgia Institute of Technology, GA·Computer Science·Early-career
Build a cross-platform framework that measures how thermal hotspots, frequency throttling, power spikes and energy limits restrict distributed AI training performance on GPUs and other accelerators.
The school
Boise State University: 10 labs on PI Pages, $58M a year on research, 64 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.