event

Ph.D. Proposal Oral Exam - Akshat Ramachandran

Title:  Co-Designing Hardware And Algorithms For Efficient Foundational Model Computing Via Adaptable Sparsity And Quantization

Committee:

Dr. Krishna, Advisor

Dr. Lin, Chair

Dr. Qureshi

Media

No media selected

Summary

The objective of this research is to enable efficient and accurate foundation model inference through cross-layer co-optimization of algorithms and hardware across a continuum of granularities—from static, model-centric strategies to dynamic, token-centric execution—leveraging flexible sparsity and quantization techniques.

Details

Thursday
Mar 19 2026
03:00pm - 05:00pm

Location: Room 1315, Klaus

In campus calendar: No

Sidebar Content

No sidebar content

Groups

Status

  • Workflow status: Published
  • Created by: Daniela Staiculescu
  • Created: 03/02/2026
  • Modified By: Daniela Staiculescu
  • Modified: 03/02/2026

Related Links

Categories

Keywords

User Data

Target Audience