Security Hardening

Security hardening focuses on enhancing the robustness and resilience of systems, particularly software and models, against attacks or failures. Current research emphasizes developing model-level hardening techniques, such as integrating error correction directly into neural networks or using diverse populations of models to improve resilience against adversarial attacks, and employing optimization algorithms like evolutionary diversity optimization to find optimal defensive strategies. These advancements are crucial for ensuring the reliability and safety of increasingly complex systems in critical applications, ranging from power grids to AI-driven decision-making.

Papers

May 17, 2024

Cost-Effective Fault Tolerance for CNNs Using Parameter Vulnerability Based Hardening and Pruning
Mohammad Hasan Ahmadilivani, Seyedhamidreza Mousavi, Jaan Raik, Masoud Daneshtalab, Maksim Jenihhin
Convolutional Neural Network Edge Pruning Fault Tolerance Security Hardening Parallel Redundancy

April 8, 2023

Evolving Reinforcement Learning Environment to Minimize Learner's Achievable Reward: An Application on Hardening Active Directory Systems
Diksha Goel, Aneta Neumann, Frank Neumann, Hung Nguyen, Mingyu Guo
Reinforcement Learning Application Proficiency Attack Graph RL Agent Defense Strategy B Learner Evolutionary Diversity Security Hardening

February 10, 2023

Large Language Models for Code: Security Hardening and Adversarial Testing
Jingxuan He, Martin Vechev
Code Generation Real World Code Adversarial Testing Code Security Security Hardening

July 5, 2022

Approximating Discontinuous Nash Equilibrial Values of Two-Player General-Sum Differential Games
Lei Zhang, Mukesh Ghimire, Wenlong Zhang, Zhe Xu, Yi Ren
Nash Equilibrium Approximate Nash Equilibrium Differential Game Equilibrium Policy Security Hardening

May 20, 2022

Multi-stage Resilience Management of Smart Power Distribution Systems: A Stochastic Robust Optimization Model
Nariman L. Dehghani, Abdollah Shafieezadeh
Robust Optimization Resilience Measure Stochastic Robust Optimization Security Hardening

November 19, 2021

Resilience from Diversity: Population-based approach to harden models against adversarial attacks
Jasser Jasser, Ivan Garibay
Adversarial Attack Adversarial Training Diversity Awareness Adversarial Sample Tiny Refinement Elicit Resilience Traditional Deep Learning Linkage Model Population Level Security Hardening

Security Hardening

Papers

Cost-Effective Fault Tolerance for CNNs Using Parameter Vulnerability Based Hardening and Pruning

Evolving Reinforcement Learning Environment to Minimize Learner's Achievable Reward: An Application on Hardening Active Directory Systems

Large Language Models for Code: Security Hardening and Adversarial Testing

Approximating Discontinuous Nash Equilibrial Values of Two-Player General-Sum Differential Games

Multi-stage Resilience Management of Smart Power Distribution Systems: A Stochastic Robust Optimization Model

Resilience from Diversity: Population-based approach to harden models against adversarial attacks