Cyan Subhra Mishra

Ph.D. Candidate in Computer Science and Engineering

About

I am a Ph.D. Candidate in the Department of Computer Science and Engineering at Pennsylvania State University, working with Dr. Mahmut Taylan Kandemir and Dr. Jack Sampson. My research focuses on hardware/software co-design for machine learning systems, with particular emphasis on resource-constrained and energy-harvesting environments.

I specialize in accelerator architecture, kernel optimization, and performance modeling for ML workloads across heterogeneous compute platforms. My work spans computer architecture, machine learning systems, and sustainable computing, with a focus on designing energy-efficient systems for edge AI applications, intermittent computing, and continuous learning on resource-constrained devices.

Research Interests

Computer Architecture for ML Sustainable Computing & Green AI Edge Computing & Energy Harvesting Intermittent Computing Hardware-Software Co-design Continuous Learning on Edge Cloud Computing & Serverless Point Cloud Processing

Education

Ph.D. in Computer Science and Engineering

Pennsylvania State University (2018-2025, Expected)

Advisors: Dr. Mahmut Taylan Kandemir, Dr. Jack Sampson

B.Tech. + M.Tech. Dual Degree in Electronics and Communication Engineering

National Institute of Technology, Rourkela (2011-2016, Honors)

CGPA: 8.39/10.00

Recent News

  • October 01, 2023
    Paper accepted at HPCA 2023
    Our paper “ResiRCA: Resilient Root Cause Analysis for Distributed Systems using Machine Learning” has been accepted at the IEEE International Symposium on High-Performance Computer Architecture (HPCA) 2023. This work presents...
  • January 01, 2025
    Paper accepted at ICLR 2025
    I’m excited to announce that our paper “Origin: On-Device Reinforcement Learning Framework for Edge Intelligence” has been accepted at the International Conference on Learning Representations (ICLR) 2025. This work introduces...
  • February 01, 2025
    Paper accepted at IPDPS 2025
    Our paper “Sustainable Edge Computing: Energy-Efficient Deep Learning Inference on Resource-Constrained Devices” has been accepted at the IEEE International Parallel and Distributed Processing Symposium (IPDPS) 2025. This work presents novel...

See all news →