2021 Adversarial Machine Learning Rising Star Award Presentation by Huan Zhang at AdvML workshop
Presentation of 2021 Adversarial Machine Learning Rising Star Award by Huan Zhang (https://www.huan-zhang.com/
More details about the workshop and AdvML Rising Star Award: https://sites.google.com/view/advml
---
Talk Title: Unboxing the Black-box: A Quest for Scalable and Powerful Neural Network Verifiers
Abstract: Neural networks have become a crucial element in modern artificial intelligence. However, they are often black-boxes and can behave unexpectedly and produce surprisingly wrong results under slightly altered inputs. When applying neural networks to mission-critical systems such as autonomous driving and aircraft control, it is often desirable to formally verify that a neural network satisfies given properties such as safety and robustness. Unfortunately, the complexity of neural networks has made the task of formally verifying their properties very challenging. To tackle this challenge, I first propose an efficient perturbation analysis algorithm based on linear relaxations of neural networks, which produces guaranteed output bounds given bounded input perturbations. The algorithm propagates linear inequalities through the network efficiently, analogous to the forward and backward propagation, and can be applied to arbitrary network architectures. To reduce relaxation error, I develop an efficient optimization procedure that can tighten linear relaxations rapidly on machine learning accelerators such as GPUs. This allows us to build a verifier which scales to much larger networks compared to existing linear programming based ones while producing tighter results. Lastly, I discuss how to further empower the verifier with branch and bound by incorporating the additional branching constraints into the bound propagation procedure. The combination of these advanced neural network verification techniques leads to α,β-CROWN, a scalable and powerful neural network verifier that is up to 2 to 3 orders of magnitudes faster than traditional CPU based neural network verifier and won the 2nd International Verification of Neural Networks Competition (VNN-COMP’21) with the highest total score over a set of 9 benchmarks.
Видео 2021 Adversarial Machine Learning Rising Star Award Presentation by Huan Zhang at AdvML workshop канала TrustworthyAI
More details about the workshop and AdvML Rising Star Award: https://sites.google.com/view/advml
---
Talk Title: Unboxing the Black-box: A Quest for Scalable and Powerful Neural Network Verifiers
Abstract: Neural networks have become a crucial element in modern artificial intelligence. However, they are often black-boxes and can behave unexpectedly and produce surprisingly wrong results under slightly altered inputs. When applying neural networks to mission-critical systems such as autonomous driving and aircraft control, it is often desirable to formally verify that a neural network satisfies given properties such as safety and robustness. Unfortunately, the complexity of neural networks has made the task of formally verifying their properties very challenging. To tackle this challenge, I first propose an efficient perturbation analysis algorithm based on linear relaxations of neural networks, which produces guaranteed output bounds given bounded input perturbations. The algorithm propagates linear inequalities through the network efficiently, analogous to the forward and backward propagation, and can be applied to arbitrary network architectures. To reduce relaxation error, I develop an efficient optimization procedure that can tighten linear relaxations rapidly on machine learning accelerators such as GPUs. This allows us to build a verifier which scales to much larger networks compared to existing linear programming based ones while producing tighter results. Lastly, I discuss how to further empower the verifier with branch and bound by incorporating the additional branching constraints into the bound propagation procedure. The combination of these advanced neural network verification techniques leads to α,β-CROWN, a scalable and powerful neural network verifier that is up to 2 to 3 orders of magnitudes faster than traditional CPU based neural network verifier and won the 2nd International Verification of Neural Networks Competition (VNN-COMP’21) with the highest total score over a set of 9 benchmarks.
Видео 2021 Adversarial Machine Learning Rising Star Award Presentation by Huan Zhang at AdvML workshop канала TrustworthyAI
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
2022 AdvML workshop Best Paper Award Presentation by Lei XuAAAI'22 Tutorial on "Adversarial Machine Learning for Good", presented by Pin-Yu Chen @ IBM Research2022 Adversarial Machine Learning Rising Star Award Presentation by Fatemehsadat MireshghallahPresentations at 3rd Workshop on Adversarial Learning Methods for Machine Learning and Data Mining[AAAI 2024] Model Reprogramming: Resource-efficient Cross-Domain Machine Learning by Pin-Yu ChenCrossFL-2022: Workshop on Cross-Community Federated Learning: Algorithms, Systems and Co-designsAAAI 2023: Holistic Adversarial Robustness of Deep Learning Models by Pin-Yu Chen2022 Adversarial Machine Learning Rising Star Award Presentation by Linyi Li at AdvML workshopCVPR 2021 Tutorial on "Practical Adversarial Robustness in Deep Learning: Problems and Solutions"CVPR 2020 Tutorial on Zeroth Order Optimization: Theory and Applications to Deep LearningECCV 2020 Tutorial on Adversarial Robustness of Deep Learning Models by Pin-Yu Chen (IBM Research)2021 Adversarial Machine Learning Rising Star Award Presentation by Florian Tramer at AdvML workshop