Socially Responsible Machine Learning: Security, Robustness, and Beyond
Explore socially responsible machine learning, focusing on security and robustness. Discover methods to enhance adversarial robustness and address challenges in the era of foundational models.
Deep Learning Robustness Verification for Few-Pixel Attacks
Explore a novel L0 robustness verifier for neural networks, addressing few-pixel attacks through dynamic programming, sampling, and covering designs to enhance verification efficiency.
Neural Nets for NLP 2021 - Adversarial Methods
Explore generative adversarial networks, adversarial feature learning, and adversarial attacks in NLP. Learn cutting-edge techniques for robust and domain-invariant neural network models in natural language processing.
Universal 3-Dimensional Perturbations for Black-Box Attacks on Video Recognition Systems
Explore innovative techniques for creating universal 3D perturbations to challenge and test the robustness of video recognition systems in black-box attack scenarios.
On the Robustness of Deep K-Nearest Neighbors
Explore the robustness of Deep k-Nearest Neighbors against adversarial attacks, proposing a heuristic method to find vulnerabilities in this defense mechanism for deep learning models.
HopSkipJumpAttack - A Query-Efficient Decision-Based Attack
Explore a query-efficient decision-based adversarial attack algorithm for generating adversarial examples, analyzing its theoretical foundations and performance against defense mechanisms.
Flawed Machine Learning Security - Alejandro Saucedo
Explore critical security flaws in ML lifecycle and learn prevention strategies. Gain insights on MLSecOps, challenges in production systems, and OWASP Top 10 for Machine Learning from an industry expert.
Building Realistic Machine Learning Systems for Security
Explore building robust machine learning systems for malware detection, addressing challenges like low false positive rates, polluted data, and realistic adversarial attacks in security applications.
Classifiers Under Attack: Evasion Techniques and Defensive Strategies
Exploring vulnerabilities in machine learning classifiers for security applications, demonstrating evasion techniques, and discussing methods to enhance robustness against adversarial attacks.
Adversarial Examples in Machine Learning - Crafting and Defending Against Attacks
Explore machine learning vulnerabilities, adversarial examples, and their impact on system security. Learn crafting techniques and defenses through practical tutorials and real-world applications.
Manifold Mixup: Better Representations by Interpolating Hidden States
Explore Manifold Mixup, a regularization technique that improves neural network performance by interpolating hidden representations, leading to smoother decision boundaries and more robust predictions.
Towards Deep Learning Models Resistant to Adversarial Attacks
Explore deep learning model vulnerabilities and defense strategies against adversarial attacks, examining research findings and experimental results.
Lessons Learned from Evaluating the Robustness of Defenses to Adversarial Examples
Explore robustness of AI defenses against adversarial examples, evaluating threat models and defense strategies to improve security in deep learning systems.
Provable Robustness Beyond Bound Propagation
Explore adversarial robustness in deep learning, focusing on randomized smoothing as a defense strategy and its theoretical guarantees for certified robustness.
Attacking Optical Flow
Explores adversarial patch attacks on optical flow networks, revealing vulnerabilities in deep learning-based motion estimation crucial for autonomous driving safety.
Practical Adversarial Attack Against Speech Recognition Platforms
Explore techniques for adversarial attacks on speech recognition systems, focusing on Amazon Echo and Kodi. Learn about blackbox approaches and philosophical implications.
Flawed ML Security: Mitigating Security Vulnerabilities in Data and Machine Learning Infrastructure with MLSecOps
Explore security vulnerabilities in machine learning infrastructure and learn practical strategies for mitigating risks using MLSecOps. Gain insights on RBAC, encryption, and best practices for robust ML systems.
Trusted and Responsible AI - Explainability, Adversarial Attacks, Bias, and Fairness
Explore key aspects of responsible AI: explainability, adversarial challenges, bias mitigation, and fairness. Learn strategies for building trustworthy and ethical AI systems in real-world applications.
Security Vulnerabilities in AI Assistant Based Applications
Explores security vulnerabilities in AI assistants, covering physical attacks, adversarial techniques, and defense strategies. Provides insights for developers and security professionals.
Security of Machine Learning Systems
Explore security challenges in AI and ML systems, including adversarial attacks and MLOps. Learn to apply pen testing skills to ML, with demonstrations on manipulating image recognition and NLP models.
Poisoned Pickles - Security Risks and Protections for Serialized ML Models
Explore the security risks of using pickle for ML model serialization and learn strategies to protect against poisoned pickles in data science pipelines.