CV

Education

• Ph.D in Computer Engineering, North Carolina State University, 2026 (expected)
• B.S.E. in Electrical and Computer Engineering, B.S. in Computer Science, Duke University, 2021

Research experience

• Compression Algorithms on AI Accelerators
  • Timeline: May 2023-Present
  • Institution: North Carolina State University and Argonne National Laboratory
  • Core Duties: Developing training and activation data compression algorithms tuned for a variety of AI accelerators including SambaNova, Cerebras, Graphcore, and Groq
  • Supervisors: Michela Becchi, Franck Cappello, Sheng Di, Xiaodong Yu
• Huffman Coding on GPU
  • Timeline: Aug 2022-Jun 2023
  • Institution: North Carolina State University and Argonne National Laboratory
  • Core Duties: Implemented and evaluated pre-computed Huffman codebook scheme to offload codebook computation from GPU using CUDA speeding up compression up to 5x and improves speed for HDF5 chunking and MPI communication.
  • Supervisors: Michela Becchi, Franck Cappello, Sheng Di, Xiaodong Yu
• GPU-based Compression for Quantum Circuit Simulations
  • Timeline: May 2022-May 2023
  • Institution: North Carolina State University and Argonne National Laboratory
  • Core Duties: Designed a lossy compression framework targeting tensors in quantum circuit tensor network-based simulation improving compression ratio relative to state-of-the-art compressors up to 10x while creating an alternate design that achieves 3-4x higher compression with limited impact on throughput
  • Supervisors: Michela Becchi, Franck Cappello, Sheng Di, Xiaodong Yu
• Accelerating Random Forest Inference on GPU
  • Timeline: Aug 2021-Apr 2022
  • Institution: North Carolina State University
  • Core Duties: Developed a novel GPU and FPGA memory layout and kernel for Random Forest algorithm to accelerate inference time using CUDA, achieving nearly 2x speedup over NVIDIA’s cuML library.
  • Supervisors: Michela Becchi
• Autoencoder for Plant Phosphoproteomics
  • Timeline: May 2021-Aug 2021
  • Institution: North Carolina State University
  • Core Duties: Optimized an autoencoder coupled with a supervised neural network for plant phenotype prediction using phosphoproteomics data. Developed a Shannon Entropy model for significant biomarker discrimination.
  • Supervisors: Ross Sozzani, Lisa Van Den Broeck
• Accelerating Semi-Supervised Support Vector Machine on FPGA
  • Timeline: Aug 2020-Dec 2021
  • Institution: Duke University
  • Core Duties: Using a Xilinx FPGA, wrote a Verilog implementation for hardware acceleration of a semi-supervised SVM. Leveraged existing mixed integer linear programming algorithms to port to FPGA.
  • Supervisors: John Board
• Image Analysis of Rice Roots
  • Timeline: Sep 2019-May 2021
  • Institution: Duke University
  • Core Duties: Developed object detection neural networks to identify phenotypic variations in roots of rice strains. Assisted in pipelining image analysis and image acquisition through cloud computing.
  • Supervisors: Philip Benfey, Isaiah Taylor
• Machine Learning Model for Biomarker Selection and Cancer Outcomes
  • Timeline: May 2019-Aug 2019
  • Institution: University of Central Florida
  • Core Duties: Assisted in developing a network-transfer and bipartite graph machine learning model to select significant biomarkers and predict cancer outcomes given genomics data, evaluated against RF and SVM.
  • Supervisors: Wei Zhang

Skills

• Programming/Hardware Languages:
  • C/C++
  • Python
  • Verilog
  • Java
• Frameworks/Tools:
  • CUDA
  • PyTorch
  • OpenMP/MPI
  • Scikit/OpenCV
  • Git
  • Vivado/Quartus

Publications

A Portable, Fast, DCT-based Compressor for AI Accelerators

Milan Shah, Xiaodong Yu, Sheng Di, Michela Becchi, and Franck Cappello. 2024. A Portable, Fast, DCT-based Compressor for AI Accelerators. In Proceedings of the 33rd International Symposium on High-Performance Parallel and Distributed Computing (HPDC '24). Association for Computing Machinery, New York, NY, USA, 109–121. https://doi.org/10.1145/3625549.3658662

The rhizodynamics robot: Automated imaging system for studying long-term dynamic root growth

Rajanala A, Taylor IW, McCaskey E, Pierce C, Ligon J, et al. (2023) The rhizodynamics robot: Automated imaging system for studying long-term dynamic root growth. PLOS ONE 18(12): e0295823. https://doi.org/10.1371/journal.pone.0295823

GPU-Accelerated Error-Bounded Compression Framework for Quantum Circuit Simulations

M. Shah et al., "GPU-Accelerated Error-Bounded Compression Framework for Quantum Circuit Simulations," 2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS), St. Petersburg, FL, USA, 2023, pp. 757-767, doi: 10.1109/IPDPS54959.2023.00081.

Lightweight Huffman Coding for Efficient GPU Compression

Milan Shah, Xiaodong Yu, Sheng Di, Michela Becchi, and Franck Cappello. 2023. Lightweight Huffman Coding for Efficient GPU Compression. In Proceedings of the 37th ACM International Conference on Supercomputing (ICS '23). Association for Computing Machinery, New York, NY, USA, 99–110. https://doi.org/10.1145/3577193.3593736

Accelerating Random Forest Classification on GPU and FPGA

Milan Shah, Reece Neff, Hancheng Wu, Marco Minutoli, Antonino Tumeo, and Michela Becchi. 2023. Accelerating Random Forest Classification on GPU and FPGA. In Proceedings of the 51st International Conference on Parallel Processing (ICPP '22). Association for Computing Machinery, New York, NY, USA, Article 4, 1–11. https://doi.org/10.1145/3545008.3545067

Network-based multi-task learning models for biomarker selection and cancer outcome prediction

Zhibo Wang, Zhezhi He, Milan Shah, Teng Zhang, Deliang Fan, Wei Zhang, Network-based multi-task learning models for biomarker selection and cancer outcome prediction, Bioinformatics, Volume 36, Issue 6, March 2020, Pages 1814–1822, https://doi.org/10.1093/bioinformatics/btz809

Posters

• Compressing Quantum Circuit Simulation Tensor Data (Best Graduate Poster Finalist)
  • International Conference for High Performance Computing, Networking, Storage, and Analysis (SC 2022)

Awards

• ACM SRC Graduate Poster Second Place (2022)
• Provost’s Doctoral Fellowship (2021)
• Graduate Merit Award (2021)
• Dean’s List-Pratt School of Engineering (x3; 2017-2020)