Dissimilarity Representation in Complex Geometric Settings

2026

A short overview of an experimental framework for benchmarking dissimilarity-based ML methods.

During a machine learning internship at the Universita di Parma, Michele built an experimental framework to compare Dissimilarity Representation with traditional feature-based methods.

The work used automated Python notebooks, Papermill, Apptainer containers, SLURM scheduling, and the university HPC cluster to run a systematic analysis across 11 datasets.