Alert button
Picture for Simone Bombari

Simone Bombari

Alert button

Towards Understanding the Word Sensitivity of Attention Layers: A Study via Random Features

Feb 05, 2024
Simone Bombari, Marco Mondelli

Viaarxiv icon

Stability, Generalization and Privacy: Precise Analysis for Random and NTK Features

May 20, 2023
Simone Bombari, Marco Mondelli

Figure 1 for Stability, Generalization and Privacy: Precise Analysis for Random and NTK Features
Figure 2 for Stability, Generalization and Privacy: Precise Analysis for Random and NTK Features
Figure 3 for Stability, Generalization and Privacy: Precise Analysis for Random and NTK Features
Figure 4 for Stability, Generalization and Privacy: Precise Analysis for Random and NTK Features
Viaarxiv icon

Beyond the Universal Law of Robustness: Sharper Laws for Random Features and Neural Tangent Kernels

Feb 03, 2023
Simone Bombari, Shayan Kiyani, Marco Mondelli

Figure 1 for Beyond the Universal Law of Robustness: Sharper Laws for Random Features and Neural Tangent Kernels
Figure 2 for Beyond the Universal Law of Robustness: Sharper Laws for Random Features and Neural Tangent Kernels
Figure 3 for Beyond the Universal Law of Robustness: Sharper Laws for Random Features and Neural Tangent Kernels
Figure 4 for Beyond the Universal Law of Robustness: Sharper Laws for Random Features and Neural Tangent Kernels
Viaarxiv icon

Memorization and Optimization in Deep Neural Networks with Minimum Over-parameterization

May 20, 2022
Simone Bombari, Mohammad Hossein Amani, Marco Mondelli

Figure 1 for Memorization and Optimization in Deep Neural Networks with Minimum Over-parameterization
Figure 2 for Memorization and Optimization in Deep Neural Networks with Minimum Over-parameterization
Viaarxiv icon

Sharp asymptotics on the compression of two-layer neural networks

May 18, 2022
Mohammad Hossein Amani, Simone Bombari, Marco Mondelli, Rattana Pukdee, Stefano Rini

Figure 1 for Sharp asymptotics on the compression of two-layer neural networks
Figure 2 for Sharp asymptotics on the compression of two-layer neural networks
Figure 3 for Sharp asymptotics on the compression of two-layer neural networks
Figure 4 for Sharp asymptotics on the compression of two-layer neural networks
Viaarxiv icon

Towards Differential Relational Privacy and its use in Question Answering

Mar 30, 2022
Simone Bombari, Alessandro Achille, Zijian Wang, Yu-Xiang Wang, Yusheng Xie, Kunwar Yashraj Singh, Srikar Appalaraju, Vijay Mahadevan, Stefano Soatto

Figure 1 for Towards Differential Relational Privacy and its use in Question Answering
Figure 2 for Towards Differential Relational Privacy and its use in Question Answering
Figure 3 for Towards Differential Relational Privacy and its use in Question Answering
Figure 4 for Towards Differential Relational Privacy and its use in Question Answering
Viaarxiv icon