Ananda Samajdar

Georgia Institute of Technology

Papers

3

Total Citations

487

H-Index

3

About

Ananda Samajdar is a leading researcher in computer architecture, with a focus on designing specialized hardware accelerators for deep learning. His work tackles the critical challenge of efficiently executing the sparse and irregular computations that underpin modern neural network training. Samajdar’s most impactful contribution is the SIGMA accelerator, a groundbreaking architecture that introduces flexible interconnects to handle sparse and irregular General Matrix Multiply (GEMM) operations. This work, published in 2020, has garnered over 460 citations, reflecting its significant influence on the field of efficient DNN training hardware. Beyond accelerating static models, Samajdar has also explored the frontier of lifelong learning with GeneSys, a hardware-software system designed to enable neural network evolution and continuous learning directly on-chip. By addressing both the demands of today’s dense models and the emerging needs of adaptive, evolving networks, Samajdar’s research provides a comprehensive vision for the future of deep learning acceleration, making him a key figure in the ongoing evolution of AI hardware.

Research Focus

Key Achievements

3
H-Index
3
Papers
487
Total Citations
162
Avg Citations/Paper
🏆 Most Cited Paper
SIGMA: A Sparse and Irregular GEMM Accelerator with Flexible Interconnects for DNN Training
462 citations · 2020
📈 Most Prolific Year: 2018 (2 Papers)
🤝 Key Collaborators: 9
🏛 Institutions: Georgia Institute of Technology

Top Papers

  1. 1
  2. 2
  3. 3

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago