Shahriar Iqbal

Shahriar Iqbal

CS PhD Candidate @USC

Shahriar Iqbal

I am a PhD candidate in the Computer Science and Engineering department at University of South Carolina Columbia SC. My current research is focused on performance optimization and debugging of highly configurable software systems. I am currently working at AISys Lab under supervision of Dr. Pooyan Jamshidi.

Prior to joining PhD, I worked as a System Software Engineer for three years at Hewlett Packard Enterprise Houston TX. I completed my Masters in Electrical Engineering from University of Central Florida Orlando FL and Bachelors in Applied Physics, Electronics and Comunication Engineering from University of Dhaka Bangladesh.

Research Interests

I am interested to pursue research in the area of systems performance, deep learning compilers, distributed systems, and causal inference. Please do not hesitate to reach out to me at miqbal@email.sc.edu if you are looking for potential collaboration opportunities.

Current Status

I am actively looking for Postdoc and Research Scientist positions.

Selected Publications

  • [SoCC’23] - CAMEO: A Causal Transfer Learning Approach for Performance Optimization of Configurable Computer Systems, Santa Cruz, CA. MS Iqbal, Z Zhong, I Ahmad, B Ray, P Jamshidi. Acceptance Rate: 20%.
  • [JAIR] - FlexiBO: A Decoupled Cost-Aware Multi-Objective Optimization Approach for Deep Neural Networks. MS Iqbal, J Su, L Kotthoff, P Jamshidi. Acceptance Rate: 24%.
  • [AutoML’22] - Getting the Best Bang For Your Buck: Choosing What to Evaluate for Faster Bayesian Optimization, Baltimore, MD. MS Iqbal, J Su, L Kotthoff, P Jamshidi.
  • [EuroSys’22] - Unicorn: Reasoning about Configurable System Performance through the lens of Causality, Rennes, France. MS Iqbal, R Krishna, MA Javidian, B Ray, P Jamshidi. Acceptance Rate: 24%.
  • [MLforSys@NeurIPS’20] - CADET: A Systematic Method For Debugging Misconfigurations using Counterfactual Reasoning, Zoomville. MS Iqbal, R Krishna, MA Javidian, B Ray, P Jamshidi.

What’s Happening?

  • Nov’23 - Successfully completed my PhD dissertation defense.
  • Oct’23 - Presented CAMEO in SoCC’23.
  • Sep’23 - Started work on LLM compiler optimization.
  • Sep’23 - Our paper CAMEO got accepted in SoCC’23.
  • Apr’23 - Our paper FlexiBO got accepted in JAIR.
  • Aug’22 - Presented our Decoupled Optimization paper at AutoML’22.
  • Jan’23 - Successfully passed PhD proposal.
  • May’22 - Our Decoupled Optimization paper got accepted into AutoML’22.
  • Feb’22 - Our Change Point Detection paper got accepted into ICPE’22.
  • Jan’22 - Our paper Unicorn got accepted in EuroSys’22.
  • Dec’20 - Presented CADET in MLforSys@NeurIPS’20.
  • Jan’21 - Presented FlexiBO to FOSD Online Meeting.
  • Nov’20 - Presented FlexiBO to ABLE group at CMU.
  • Oct’20 - Our paper CADET got accepted in MLforSys@NeurIPS’20.
  • May’19 - Presented our Performance Modeling paper got accepted in USENIX OpML’19.
  • May’19 - Our paper on Performance Modeling got accepted in USENIX OpML’19.