Zobrazeno 1 - 10
of 1 584
pro vyhledávání: '"Biggio P"'
Autor:
Ledda, Emanuele, Scodeller, Giovanni, Angioni, Daniele, Piras, Giorgio, Cinà, Antonio Emanuele, Fumera, Giorgio, Biggio, Battista, Roli, Fabio
In learning problems, the noise inherent to the task at hand hinders the possibility to infer without a certain degree of uncertainty. Quantifying this uncertainty, regardless of its wide use, assumes high relevance for security-sensitive application
Externí odkaz:
http://arxiv.org/abs/2410.21952
Current progress in artificial intelligence is centered around so-called large language models that consist of neural networks processing long sequences of high-dimensional vectors called tokens. Statistical physics provides powerful tools to study t
Externí odkaz:
http://arxiv.org/abs/2410.18858
Autor:
Piras, Giorgio, Pintor, Maura, Demontis, Ambra, Biggio, Battista, Giacinto, Giorgio, Roli, Fabio
Recent work has proposed neural network pruning techniques to reduce the size of a network while preserving robustness against adversarial examples, i.e., well-crafted inputs inducing a misclassification. These methods, which we refer to as adversari
Externí odkaz:
http://arxiv.org/abs/2409.01249
Autor:
Villani, Francesco, Lazzaro, Dario, Cinà, Antonio Emanuele, Dell'Amico, Matteo, Biggio, Battista, Roli, Fabio
Data poisoning attacks on clustering algorithms have received limited attention, with existing methods struggling to scale efficiently as dataset sizes and feature counts increase. These attacks typically require re-clustering the entire dataset mult
Externí odkaz:
http://arxiv.org/abs/2408.07558
How do different architectural design choices influence the space of solutions that a transformer can implement and learn? How do different components interact with each other to shape the model's hypothesis space? We investigate these questions by c
Externí odkaz:
http://arxiv.org/abs/2407.11542
Autor:
Mura, Raffaele, Floris, Giuseppe, Scionis, Luca, Piras, Giorgio, Pintor, Maura, Demontis, Ambra, Giacinto, Giorgio, Biggio, Battista, Roli, Fabio
Gradient-based attacks are a primary tool to evaluate robustness of machine-learning models. However, many attacks tend to provide overly-optimistic evaluations as they use fixed loss functions, optimizers, step-size schedulers, and default hyperpara
Externí odkaz:
http://arxiv.org/abs/2407.08806
Autor:
Scano, Christian, Floris, Giuseppe, Montaruli, Biagio, Demetrio, Luca, Valenza, Andrea, Compagna, Luca, Ariu, Davide, Piras, Luca, Balzarotti, Davide, Biggio, Battista
ModSecurity is widely recognized as the standard open-source Web Application Firewall (WAF), maintained by the OWASP Foundation. It detects malicious requests by matching them against the Core Rule Set (CRS), identifying well-known attack patterns. E
Externí odkaz:
http://arxiv.org/abs/2406.13547
Autor:
Chen, Zhang, Demetrio, Luca, Gupta, Srishti, Feng, Xiaoyi, Xia, Zhaoqiang, Cinà, Antonio Emanuele, Pintor, Maura, Oneto, Luca, Demontis, Ambra, Biggio, Battista, Roli, Fabio
Thanks to their extensive capacity, over-parameterized neural networks exhibit superior predictive capabilities and generalization. However, having a large parameter space is considered one of the main suspects of the neural networks' vulnerability t
Externí odkaz:
http://arxiv.org/abs/2406.10090
Autor:
Ponte, Andrea, Trizna, Dmitrijs, Demetrio, Luca, Biggio, Battista, Ogbu, Ivan Tesfai, Roli, Fabio
As a result of decades of research, Windows malware detection is approached through a plethora of techniques. However, there is an ongoing mismatch between academia -- which pursues an optimal performances in terms of detection rate and low false ala
Externí odkaz:
http://arxiv.org/abs/2405.14478
Deep learning-based malware detection systems are vulnerable to adversarial EXEmples - carefully-crafted malicious programs that evade detection with minimal perturbation. As such, the community is dedicating effort to develop mechanisms to defend ag
Externí odkaz:
http://arxiv.org/abs/2405.00392