← Return to homepage

Pavan Kumar MP, Ph.D.

Dhahran, Saudi Arabia · pavanforyou94@gmail.com · GitHub · LinkedIn

Professional Experience

Postdoctoral Researcher — KFUPM

November 2025–Present · Dhahran, Saudi Arabia

• Physics-conditioned LLMs and LoRA/QLoRA for interpretable industrial fault diagnosis.
• Battery state-of-health and remaining useful life (RUL) prediction; turbofan RUL modeling.
• Evaluation of OpenVLA for robotic manipulation and industrial automation.

ML Research Intern — National Center for High-Performance Computing

January–May 2024 · Taiwan

Physics-informed neural networks using NVIDIA Modulus Sym and scientific machine learning.

PCB Check-in Engineer — Sierra Circuits

October 2016–October 2020 · Bangalore, India

PCB fabrication feasibility, client requirements, and manufacturing workflow review.

Education

• Ph.D., Computer Science Engineering, National Sun Yat-sen University, Taiwan (2020–2024). Advisor: Prof. Kun-Chih Chen.
• M.Tech., VLSI Design, Visvesvaraya Technological University, India (2017–2019).
• B.E., Electronics and Communication, Visvesvaraya Technological University, India (2012–2016).

Research and Skills

Predictive maintenance, fault diagnosis, unsupervised domain adaptation, physics-informed AI, LLMs, LoRA/QLoRA, battery SoH, RUL estimation. Python, R, Verilog, PyTorch, TensorFlow, Hugging Face Transformers, NVIDIA Modulus Sym, scikit-learn, Pandas, NumPy, OpenCV.

Selected Journal Publications

1. Neural Network-Based Anomaly Detection Architecture for Bearing Fault Diagnosis. IEEE TIM (2025). DOI.
2. Mitigating Negative Transfer Learning in Source-Free Unsupervised Domain Adaptation for Rotating Machinery Fault Diagnosis. IEEE TIM (2024). DOI.
3. Enhancing Learning in Fine-Tuned Transfer Learning for Rotating Machinery via Negative Transfer Mitigation. IEEE TIM (2024). DOI.
4. Time Series-Based Sensor Selection and Lightweight Neural Architecture Search for RUL Estimation in Future Industry 4.0. IEEE JETCAS (2023). DOI.

Selected Conference Publications

• PC-LLM: Physics-Conditioned Large Language Models for Interpretable Bearing Fault Diagnosis in Industry 4.0. IEEE CASE (2026).
• Fine-Tuned Based Transfer Learning with Temporal Attention and Physics-Informed Loss. IEEE AICAS (2024).
• Two bearing fault diagnosis papers. IEEE COINS (2023).

Awards

VTU 4th Rank Award and Gold Medal (2019); Elite Ph.D. Scholarship, NSYSU (2020).

Public professional CV. Personal and referees' phone numbers intentionally omitted.