Support-Vector Machine è un modello di machine-learning per problemi di classificazione. Riscopriamo insieme a Luca Ciuffreda di Prometeia il paper originale del 1995. La presentazione sarà in lingua italiana.

Short Bio. Luca Ciuffreda è attualmente Data Scientist presso Prometeia dove, all'interno della divisione Data Science, contribuisce a sviluppare progetti di AI in ambito finanziario, bancario e assicurativo. Precedentemente ha lavorato presso istituti di ricerca quali SISSA e ICTP/CNR su progetti di ricerca in computational neuroscience, deep learning e computer vision, conseguendo un master in High Performance Computing.

Full-Text of the Paper. Springer (full-text available)

Original Abstract.

The support-vector network is a new learning machine for two-group classification problems. The machine conceptually implements the following idea: input vectors are non-linearly mapped to a very high-dimension feature space. In this feature space a linear decision surface is constructed. Special properties of the decision surface ensures high generalization ability of the learning machine. The idea behind the support-vector network was previously implemented for the restricted case where the training data can be separated without errors. We here extend this result to non-separable training data.

High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated. We also compare the performance of the support-vector network to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.