On the Two-fold Role of Logic Constraints in Deep Learning

Deep Learning (DL) is a branch of Artificial Intelligence (AI) that focuses on training deep neural networks. Thanks to their ability to process large amounts of data, these networks have achieved remarkable results across a variety of fields. Despite these successes, DL still faces several limitati...

Whakaahuatanga katoa

I tiakina i:
Ngā taipitopito rārangi puna kōrero
Kaituhi matua: Ciravegna, Gabriele
Hōputu: Online
Reo:Ingarihi
I whakaputaina: Firenze University Press 2025
Ngā marau:
Urunga tuihono:ONIX_20250801T172941_9791221506808_44
Ngā Tūtohu: Tāpirihia he Tūtohu
Kāore He Tūtohu, Me noho koe te mea tuatahi ki te tūtohu i tēnei pūkete!
_version_ 1869517659599536128
author Ciravegna, Gabriele
author_browse Ciravegna, Gabriele
author_facet Ciravegna, Gabriele
author_sort Ciravegna, Gabriele
collection Directory of Open Access Books
description Deep Learning (DL) is a branch of Artificial Intelligence (AI) that focuses on training deep neural networks. Thanks to their ability to process large amounts of data, these networks have achieved remarkable results across a variety of fields. Despite these successes, DL still faces several limitations that hinder its adoption in real-world scenarios. This thesis addresses three key challenges: reducing the need for supervision, defending against adversarial attacks, and explaining neural network behavior. The first two challenges are tackled through learning from constraints, which incorporates domain knowledge to guide the learning process and enhance model robustness. The third challenge, on the other hand, is addressed using learning of constraints, which helps identify and formalize logical relationships among learned tasks, thereby providing interpretable explanations of the networks’ behavior.
format Online
id doab-20.500.12854ir-163658
institution Directory of Open Access Books
language eng
publishDate 2025
publishDateRange 2025
publishDateSort 2025
publisher Firenze University Press
publisherStr Firenze University Press
record_format ojs
spelling doab-20.500.12854ir-1636582025-08-02T05:21:00Z On the Two-fold Role of Logic Constraints in Deep Learning Ciravegna, Gabriele Deep Learning (DL) Logic Constraints Active Learning Adversarial Defense Logic Explanations thema EDItEUR::U Computing and Information Technology Deep Learning (DL) is a branch of Artificial Intelligence (AI) that focuses on training deep neural networks. Thanks to their ability to process large amounts of data, these networks have achieved remarkable results across a variety of fields. Despite these successes, DL still faces several limitations that hinder its adoption in real-world scenarios. This thesis addresses three key challenges: reducing the need for supervision, defending against adversarial attacks, and explaining neural network behavior. The first two challenges are tackled through learning from constraints, which incorporates domain knowledge to guide the learning process and enhance model robustness. The third challenge, on the other hand, is addressed using learning of constraints, which helps identify and formalize logical relationships among learned tasks, thereby providing interpretable explanations of the networks’ behavior. 2025-08-02T05:20:59Z 2025-08-02T05:20:59Z 2025-08-01T15:36:14Z 2025 book ONIX_20250801T172941_9791221506808_44 https://library.oapen.org/handle/20.500.12657/104535 9791221506808 9791221506792 9791221506815 https://directory.doabooks.org/handle/20.500.12854/163658 eng Premio Tesi di Dottorato Città di Firenze open access image/jpeg Attribution 4.0 International https://library.oapen.org/bitstream/20.500.12657/104535/1/44798.pdf Firenze University Press 10.36253/979-12-215-0680-8 10.36253/979-12-215-0680-8 2ec4474d-93b1-4cfa-b313-9c6019b51b1a 9791221506808 9791221506792 9791221506815 126 Florence open access
spellingShingle Deep Learning (DL)
Logic Constraints
Active Learning
Adversarial Defense
Logic Explanations
thema EDItEUR::U Computing and Information Technology
Ciravegna, Gabriele
On the Two-fold Role of Logic Constraints in Deep Learning
title On the Two-fold Role of Logic Constraints in Deep Learning
title_full On the Two-fold Role of Logic Constraints in Deep Learning
title_fullStr On the Two-fold Role of Logic Constraints in Deep Learning
title_full_unstemmed On the Two-fold Role of Logic Constraints in Deep Learning
title_short On the Two-fold Role of Logic Constraints in Deep Learning
title_sort on the two fold role of logic constraints in deep learning
topic Deep Learning (DL)
Logic Constraints
Active Learning
Adversarial Defense
Logic Explanations
thema EDItEUR::U Computing and Information Technology
topic_facet Deep Learning (DL)
Logic Constraints
Active Learning
Adversarial Defense
Logic Explanations
thema EDItEUR::U Computing and Information Technology
url ONIX_20250801T172941_9791221506808_44
work_keys_str_mv AT ciravegnagabriele onthetwofoldroleoflogicconstraintsindeeplearning