Gabriele Martino | PhD Student

Bio

I am a PhD student in the Probabilistic and Interactive Machine Learning Group at the University of Vienna, supervised by Prof. Sebastian Tschiatschek.

I work on deep learning and explainable AI for biology, building models that can uncover patterns in nature while making their decisions easier to understand.

Interests

Deep Learning
Explainable Artificial Intelligence
Machine Learning for Biology

Current Position

PhD Student
Probabilistic and Interactive Machine Learning Group
University of Vienna

Selected Publications

RiboUnmix neural architecture and post-network equations

Gabriele Martino, Denis Skibinski, Ivo L. Hofacker, Sebastian Tschiatschek

2026 · arXiv:2609.39644

RiboUnmix: Learning Shared Translational Dynamics from Biased and Noisy Ribo-seq Measurements

A probabilistic multi-dataset framework that separates shared, sequence-dependent ribosome occupancy patterns from dataset-specific experimental distortions and stochastic measurement noise.

Preprint Code

Diagram of the FRInGe attribution path through predictive and input space

Gabriele Martino, Sebastian Tschiatschek

NeurIPS 2026 · Poster · arXiv:2605.06404

FRInGe: Distribution-Space Integrated Gradients with Fisher–Rao Geometry

A Fisher–Rao-based attribution method that defines references and interpolation in predictive distribution space, improving calibration-oriented explanation metrics across ImageNet models.

Preprint Code

Writing

Research notes, tutorials and longer technical pieces.

Visit the blog

Contact

If you would like to discuss research or collaboration, you can reach me at:

gabriele.martino [at] univie.ac.at