Portrait of Neel Gupta

Neel Gupta

Ph.D. student in Computer Engineering
University of Southern California

neeleshg@usc.edu · Google Scholar · LinkedIn · GitHub

About

I am a Ph.D. student in Computer Engineering at USC, working in the FPGA/Parallel Computing Lab, advised by Prof. Viktor K. Prasanna and Prof. Raj Kannan.

My research interests include machine learning for systems, graph machine learning, agentic and physical AI, subquadratic neural network architectures, and heterogeneous computing.

Selected Projects

Teaching

Fall 2026: Lab Teaching Assistant, EE 354L: Introduction to Digital Circuits, USC.

Service

HiPC 2025: Publicity Chair, IEEE International Conference on High Performance Computing, Data, and Analytics.

Publications

Italicized names indicate an undergraduate mentee.

  1. Topology Survives Shuffling: Robust Long-Context Memory Access Prediction for Graph Analytics [Best Student Paper]
    Dongyan Sun, Neelesh Gupta, Rajgopal Kannan, Viktor K. Prasanna
    IEEE High Performance Extreme Computing Conference (HPEC 2026).

  2. Net2Tab: Tabularizing Neural Networks with Applications to Data Prefetching
    Pengmiao Zhang, Neelesh Gupta, Rajgopal Kannan, Viktor K. Prasanna
    Journal of Parallel and Distributed Computing, vol. 213, article 105265, July 2026.
    Paper

  3. A Persistent-State Dataflow Accelerator for Memory-Bound Linear Attention Decode on FPGA [Best Poster]
    Neelesh Gupta, Peter Wang, Rajgopal Kannan, Viktor K. Prasanna
    Reconfigurable Architectures Workshop (RAW 2026), held with IEEE IPDPS, poster.
    arXiv

  4. Context-Driven Performance Modeling for Causal Inference Operators on Neural Processing Units
    Neelesh Gupta, Rakshith Jayanth, Dhruv Parikh, Viktor Prasanna
    32nd IEEE International Conference on High Performance Computing, Data, and Analytics (HiPC 2025), pp. 365–375.
    PDF

  5. Enabling Long FFT Convolutions on Memory-Constrained FPGAs via Chunking
    Peter Wang, Neelesh Gupta, Viktor K. Prasanna
    32nd IEEE International Conference on High Performance Computing, Data and Analytics Workshop (HiPCW 2025), pp. 321–322.
    PDF Paper

  6. Performance-Energy Characterization of ML Inference on Heterogeneous Edge AI Platforms
    Palash Kohli, Rakshith Jayanth, Neelesh Gupta, Haoyang Fan, Viktor Prasanna
    IEEE High Performance Extreme Computing Conference (HPEC 2025), pp. 1–7.
    Paper

  7. Towards Real-Time LLM Inference on Heterogeneous Edge Platforms
    Rakshith Jayanth, Neelesh Gupta, Souvik Kundu, Deepak A. Mathaikutty, Viktor Prasanna
    31st IEEE International Conference on High Performance Computing, Data and Analytics Workshop (HiPCW 2024), pp. 197–198.
    PDF Paper

  8. Benchmarking Edge AI Platforms for High-Performance ML Inference
    Rakshith Jayanth, Neelesh Gupta, Viktor K. Prasanna
    IEEE High Performance Extreme Computing Conference (HPEC 2024).
    PDF

  9. TabConv: Low-Computation CNN Inference via Table Lookups
    Neelesh Gupta, Narayanan Kannan, Pengmiao Zhang, Viktor K. Prasanna
    Proceedings of the 21st ACM International Conference on Computing Frontiers (CF 2024).
    PDF Code

  10. Attention, Distillation, and Tabularization: Towards Practical Neural Network-Based Prefetching
    Pengmiao Zhang, Neelesh Gupta, Rajgopal Kannan, Viktor K. Prasanna
    38th IEEE International Parallel & Distributed Processing Symposium (IPDPS 2024).
    PDF Code

  11. PaCKD: Pattern-Clustered Knowledge Distillation for Compressing Memory Access Prediction Models
    Neelesh Gupta, Pengmiao Zhang, Rajgopal Kannan, Viktor K. Prasanna
    IEEE High Performance Extreme Computing Conference (HPEC 2023).
    PDF Code

  12. Did your child get disturbed by an inappropriate advertisement on YouTube?
    Jeffrey Liu, Rajat Tandon, Uma Durairaj, Jiani Guo, Spencer Zahabizadeh, Sanjana Ilango, Jeremy Tang, Neelesh Gupta, Zoe Zhou, Jelena Mirkovic
    Proceedings of the KDD Undergraduate Consortium (KDD-UC 2022).
    PDF