Mahdi Imani

alt text 

PhD Candidate,
Department of Electrical and Computer Engineering,
Texas A&M University
College Station, Texas

Profiles: Google Scholar, ResearchGate
Email: m.imani88 [AT] tamu.edu

Education

PhD Candidate [2014 - Present]

  • Texas A&M University, Electrical Engineering

Master of Science [2012 - 2014]

  • University of Tehran, Electrical Engineering

Bachelor of Science [2007 - 2012]

  • University of Tehran, Mechanical Engineering

Research Interests

  • Machine Learning and Statistical Learning Theory

  • Statistical Pattern Recognition and Signal Processing

  • Bayesian Optimization and Statistics

  • Computational Biology and Bioinformatics

What do I do?

My research interests span the areas of machine learning, control theory and statistical signal processing. I seek to find solutions for decision making in large complex systems for fast learning, inference, data gathering and experimental design processes, enabling tackling the complex systems and possibly noisy and incomplete real-world data. In addition to decision making, I am working on developing effective tools for learning, inference and control of nonlinear/non-Gaussian state-space models. In particular, I have developed several tools for a general class of nonlinear state space models with Boolean state variables, known as partially-observed Boolean dynamical systems (POBDS). These include the optimal minimum mean-square error (MMSE) state estimators, which are called the Boolean Kalman Filter (BKF) and Boolean Kalman Smoother (BKS), adaptive filters for simultaneous estimation of state and parameters, and efficient particle filters. Finally, the last category of my research focuses on finding solutions for classification in non-stationary environments.

See some highlighted projects here.

Recent Publications

Full list of publications.