Download video

USENIX Security '16 - Stealing Machine Learning Models via Prediction APIs

207 views

USENIX

2 years ago

USENIX Security '16 - Stealing Machine Learning Models via Prediction APIs

USENIX Security '16 - Stealing Machine Learning Models via Prediction APIs

USENIX Security '22 - On the Security Risks of AutoML
10:16
USENIX Security '22 - On the Security Risks of AutoML
by USENIX
USENIX Security '23 - The Space of Adversarial Strategies
11:48
USENIX Security '23 - The Space of Adversarial Strategies
by USENIX
USENIX Security '24 - Yes, One-Bit-Flip Matters! Universal DNN Model Inference Depletion with...
12:17
USENIX Security '24 - Yes, One-Bit-Flip Matters! Universal DNN Model Inference Depletion with...
by USENIX
USENIX Security '23 - Rethinking White-Box Watermarks on Deep Learning Models under Neural...
10:17
USENIX Security '23 - Rethinking White-Box Watermarks on Deep Learning Models under Neural...
by USENIX
USENIX Security '24 - Making Them Ask and Answer: Jailbreaking Large Language Models in Few...
14:07
USENIX Security '24 - Making Them Ask and Answer: Jailbreaking Large Language Models in Few...
by USENIX
USENIX Security '22 - FLAME: Taming Backdoors in Federated Learning
10:20
USENIX Security '22 - FLAME: Taming Backdoors in Federated Learning
by USENIX
USENIX Security '24 - SecurityNet: Assessing Machine Learning Vulnerabilities on Public Models
11:42
USENIX Security '24 - SecurityNet: Assessing Machine Learning Vulnerabilities on Public Models
by USENIX
USENIX Security '24 - FIRE: Combining Multi-Stage Filtering with Taint Analysis for Scalable...
12:20
USENIX Security '24 - FIRE: Combining Multi-Stage Filtering with Taint Analysis for Scalable...
by USENIX
USENIX Security '23 - Two-in-One: A Model Hijacking Attack Against Text Generation Models
13:16
USENIX Security '23 - Two-in-One: A Model Hijacking Attack Against Text Generation Models
by USENIX
Analyzing Leakage of Personally Identifiable Information in Language Models
13:30
Analyzing Leakage of Personally Identifiable Information in Language Models
by IEEE Symposium on Security and Privacy
USENIX Security '24 - PentestGPT: Evaluating and Harnessing Large Language Models for Automated...
10:03
USENIX Security '24 - PentestGPT: Evaluating and Harnessing Large Language Models for Automated...
by USENIX
Exploiting Intelligent Systems using Model Inversion Attacks
4:59
Exploiting Intelligent Systems using Model Inversion Attacks
by Noah Scharrenberg
USENIX Security '22 - Are Your Sensitive Attributes Private? Novel Model Inversion Attribute
14:06
USENIX Security '22 - Are Your Sensitive Attributes Private? Novel Model Inversion Attribute
by USENIX
USENIX Security '24 - PatchCURE: Improving Certifiable Robustness, Model Utility, and Computation...
10:23
USENIX Security '24 - PatchCURE: Improving Certifiable Robustness, Model Utility, and Computation...
by USENIX
USENIX Security '23 - Pass2Edit: A Multi-Step Generative Model for Guessing Edited Passwords
11:19
USENIX Security '23 - Pass2Edit: A Multi-Step Generative Model for Guessing Edited Passwords
by USENIX
USENIX Security '24 - Towards More Practical Threat Models in Artificial Intelligence Security
13:37
USENIX Security '24 - Towards More Practical Threat Models in Artificial Intelligence Security
by USENIX
USENIX Security '24 - SOAP: A Social Authentication Protocol
11:45
USENIX Security '24 - SOAP: A Social Authentication Protocol
by USENIX
USENIX Security '24 - AI Psychiatry: Forensic Investigation of Deep Learning Networks in Memory...
9:15
USENIX Security '24 - AI Psychiatry: Forensic Investigation of Deep Learning Networks in Memory...
by USENIX
USENIX Security '24 - Double Face: Leveraging User Intelligence to Characterize and Recognize...
12:30
USENIX Security '24 - Double Face: Leveraging User Intelligence to Characterize and Recognize...
by USENIX
USENIX Security '23 - DiffSmooth: Certifiably Robust Learning via Diffusion Models and Local...
11:49
USENIX Security '23 - DiffSmooth: Certifiably Robust Learning via Diffusion Models and Local...
by USENIX
USENIX Security '24 - Prompt Stealing Attacks Against Text-to-Image Generation Models
12:38
USENIX Security '24 - Prompt Stealing Attacks Against Text-to-Image Generation Models
by USENIX
USENIX Security '22 - Hidden Trigger Backdoor Attack on NLP Models via Linguistic Style Manipulation
11:10
USENIX Security '22 - Hidden Trigger Backdoor Attack on NLP Models via Linguistic Style Manipulation
by USENIX
USENIX Security '24 - SoK: Neural Network Extraction Through Physical Side Channels
12:57
USENIX Security '24 - SoK: Neural Network Extraction Through Physical Side Channels
by USENIX