Reachable Polyhedral Marching (RPM): A Safety Verification Algorithm for Deep Neural Networks

Reachable Polyhedral Marching (RPM): A Safety Verification Algorithm for Deep Neural Networks

[POPL'22] Interval Universal Approximation for Neural NetworksПодробнее

[POPL'22] Interval Universal Approximation for Neural Networks

[CDC 2021] Tutorial: Neural Network Verification in ControlПодробнее

[CDC 2021] Tutorial: Neural Network Verification in Control

Towards Verification of General Neural Networks: A Dual ApproachПодробнее

Towards Verification of General Neural Networks: A Dual Approach

Methods to Soundly Verify Deep Neural Networks | Tomer Arnon | JuliaCon2021Подробнее

Methods to Soundly Verify Deep Neural Networks | Tomer Arnon | JuliaCon2021

Verification and Adaptation of Deep Neural NetworksПодробнее

Verification and Adaptation of Deep Neural Networks

FFNN Polytope-8-22-2020Подробнее

FFNN Polytope-8-22-2020

[L4DC 2022] A Simple and Efficient Sampling-based Reachability Analysis Algorithm (Hardware Results)Подробнее

[L4DC 2022] A Simple and Efficient Sampling-based Reachability Analysis Algorithm (Hardware Results)

Scaling Polyhedral Neural Network Verification on GPUs [MLSys'21]Подробнее

Scaling Polyhedral Neural Network Verification on GPUs [MLSys'21]

SoK: Certified Robustness for Deep Neural NetworksПодробнее

SoK: Certified Robustness for Deep Neural Networks

USENIX Security '18 - Formal Security Analysis of Neural Networks using Symbolic IntervalsПодробнее

USENIX Security '18 - Formal Security Analysis of Neural Networks using Symbolic Intervals

What Does It Mean to “Verify” a Neural Network?Подробнее

What Does It Mean to “Verify” a Neural Network?

Session 6: Neural Networks - Interval Methods for Resolving Neural Computation IssuesПодробнее

Session 6: Neural Networks - Interval Methods for Resolving Neural Computation Issues

Interval Universal Approximation for Neural Networks (Teaser)Подробнее

Interval Universal Approximation for Neural Networks (Teaser)

ICRA 2021 Presentation: Reachability-based Trajectory SafeguardПодробнее

ICRA 2021 Presentation: Reachability-based Trajectory Safeguard

AAAI 2022 Tutorial: "Formal Verification of Deep Neural Networks: Theory and Practice"Подробнее

AAAI 2022 Tutorial: 'Formal Verification of Deep Neural Networks: Theory and Practice'

Verifying Low-dimensional Input Neural Networks via Input QuantizationПодробнее

Verifying Low-dimensional Input Neural Networks via Input Quantization

Fluent Bit: Better Pipelines for Observability - Eduardo Silva, ChronosphereПодробнее

Fluent Bit: Better Pipelines for Observability - Eduardo Silva, Chronosphere