no code implementations • 13 Jan 2023 • Lorenzo Ciampiconi, Adam Elwood, Marco Leonardi, Ashraf Mohamed, Alessandro Rozza
This survey aims to provide a reference of the most essential loss functions for both beginner and advanced machine learning practitioners.
no code implementations • 12 Oct 2022 • Adam Elwood, Marco Leonardi, Ashraf Mohamed, Alessandro Rozza
This provides practitioners with new techniques that perform well in static and dynamic settings, and are particularly well suited to non-linear scenarios with continuous action spaces.
no code implementations • 8 Nov 2021 • Luigi Celona, Marco Leonardi, Paolo Napoletano, Alessandro Rozza
In this paper we propose a method for the automatic prediction of the aesthetics of an image that is based on the analysis of the semantic content, the artistic style and the composition of the image.
no code implementations • 29 Jul 2021 • Adam Elwood, Alberto Gasparin, Alessandro Rozza
With the rise in use of social media to promote branded products, the demand for effective influencer marketing has increased.
no code implementations • 14 Nov 2020 • Franco Manessi, Alessandro Rozza
Self-supervised learning is currently gaining a lot of attention, as it allows neural networks to learn robust representations from large quantities of unlabeled data.
no code implementations • 29 Jan 2018 • Franco Manessi, Alessandro Rozza
In the last decade, an active area of research has been devoted to design novel activation functions that are able to help deep neural networks to converge, obtaining better performance.
no code implementations • 5 Dec 2017 • Franco Manessi, Alessandro Rozza, Simone Bianco, Paolo Napoletano, Raimondo Schettini
In this work we present a method to improve the pruning step of the current state-of-the-art methodology to compress neural networks.
no code implementations • 20 Apr 2017 • Franco Manessi, Alessandro Rozza, Mario Manzo
Many different classification tasks need to manage structured data, which are usually modeled as graphs.
2 code implementations • CVPR 2017 • Giorgio Patrini, Alessandro Rozza, Aditya Menon, Richard Nock, Lizhen Qu
We present a theoretically grounded approach to train deep neural networks, including recurrent networks, subject to class-dependent label noise.
Ranked #2 on Image Classification on Clothing1M (using clean data) (using extra training data)
no code implementations • 25 Jul 2016 • Pietro Cassara, Alessandro Rozza, Mirco Nanni
From a machine learning point of view, identifying a subset of relevant features from a real data set can be useful to improve the results achieved by classification methods and to reduce their time and space complexity.