Mateusz Susik

This paper shows that variational inference methods like SVGD and VI can speed up DNA mixture deconvolution with minimal loss of accuracy.

The paper validates a Hamiltonian Monte Carlo genotyping algorithm on DNA mixtures, showing reliable likelihood ratios and proposing a new precision benchmark.

This paper introduces a Hamiltonian Monte Carlo algorithm with strict convergence criteria to reduce run-to-run variability in forensic DNA mixture deconvolution without increasing runtime.

This paper introduces an adaptive particle representation for fluorescence microscopy images to overcome storage and processing bottlenecks by replacing pixels with content-aware particles.