Abhishek Tyagi

3205 Wegmans Hall
250 Hutchison Rd
Rochester, NY 14620
I am a Fifth Year Computer Science PhD student at the University of Rochester, advised by Dr. Yuhao Zhu. I was previously a Senior Engineer at Samsung Research & Development Institute, Bangalore, India. Before that, I was a Research Assistant at CAD Lab, IISc, working with Prof SK Nandy.
I am interested in building intelligent, efficient, and reliable computing systems inspired by humans and for humans. My research philosophy is to combine my interests in understanding the world around us (for example, human brain functioning) with my expertise in the area of computer architecture and design applications and systems that facilitate what we’ve yet to understand.
news
May 19, 2025 | I have joined NVIDIA as a Resiliency and Safety Architecture Intern for the Summer and will be working with GPU Resiliency and Safety team led by Sanjay Gongalore. |
---|---|
May 2, 2025 | Our work, Dynamic Sparse Training of Diagonally Sparse Networks, has been accepted to ICML 2025 as a poster. See you in Vancouver in July! |
Apr 18, 2025 | I will be at CLEO 2025 presenting our work titled Diagonal Structured Sparsity for Efficient Optical and GPU-based Neural Networks. |
Mar 4, 2025 | 3 papers accepted for Oral presentation at Conference on Lasers and Electro-Optics (CLEO 2025). |
Jun 11, 2024 | Our work on Ethos-U55 fault characterization is featured on Arm Community Blogs: When Zero Becomes One |
selected publications
- ISPASSCharacterizing Soft-Error Resiliency in Arm’s Ethos-U55 Embedded Machine Learning AcceleratorIn 2024 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), 2024
- SIGGRAPHImperceptible Color Modulation for Power Saving in VR/ARIn ACM SIGGRAPH 2023 Emerging Technologies, 2023
- HPCAThales: Formulating and estimating architectural vulnerability factors for dnn acceleratorsIn 2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA), 2023
- SIGGRAPHColor-perception-guided display power reduction for virtual realityACM Transactions on Graphics (TOG), 2022
- ISCASTransport Triggered near Memory Accelerator for Deep LearningIn 2021 IEEE International Symposium on Circuits and Systems (ISCAS), 2021