Muhammad Izzatullah
Crafting cutting-edge deep generative models and quantifying uncertainties in waveform imaging and inversion, and more.
Bio
I am a PhD candidate at KAUST in the
Physical Science and Engineering division and affiliate with the
Seismic Wave Analysis Group,
Deep Imaging Group, and
Extreme Computing Research Center.
I specialize in deep generative modelling, Bayesian inference and uncertainty quantification.
My research contributions spanned across the fields of waveform imaging and inversion through the development
of novel deep generative modelling approach and scalable Bayesian inference methods aimed at
quantifying the uncertainties of subsurface imaging and inversion products. I am working
closely under supervision of Tariq Alkhalifah and Matteo Ravasi.
I am also actively involve with philanthropist works, and I actively support cerebral palsy, autism,
and cancer organisations in Malaysia and Russia.
Recent Updates
- Our Geophysical Journal International (GJI) paper "Muhammad Izzatullah, Tristan van Leeuwen, Daniel Peter, Bayesian seismic inversion: a fast sampling Langevin dynamics Markov chain Monte Carlo method " is now available online.
Research Activities
Deep-learning for geophysical inverse problems
Precondition physics-driven inverse problems with non-linear dimensionality reduction networks.
Novel objective functions for full waveform inversion
Use of optimal transport and deep learning to increase the sensitivity of full-waveform inversion cost functions to target areas.
