ROAST: Risk-aware Outlier-exposure for Adversarial Selective Training of Anomaly Detectors Against Evasion Attacks
ROAST: Risk-aware Outlier-exposure for Adversarial Selective Training of Anomaly Detectors Against Evasion Attacks explores A novel framework for anomaly detection that improves recall by selectively training on less vulnerable patient data and using outlier exposure to maintain precision.. Commercial viability score: 4/10 in Anomaly Detection.
BUILDER'S SANDBOX
Build This Paper
Use an AI coding agent to implement this research.
OpenAI CodexAI Agent
Lightweight coding agent in your terminal.
Claude CodeAI Agent
Agentic coding tool for terminal workflows.
AntiGravity IDEScaffolding
AI agent mindset installer and workflow scaffolder.
CursorIDE
AI-first code editor built on VS Code.
VS CodeIDE
Free, open-source editor by Microsoft.
Recommended Stack
Startup Essentials
Estimated $9K - $13K over 6-10 weeks.
See exactly what it costs to build this -- with 3 comparable funded startups.
7-day free trial. Cancel anytime.
Discover the researchers behind this paper and find similar experts.
7-day free trial. Cancel anytime.
Related Papers
- Builds On ThisNERO-Net: A Neuroevolutionary Approach for the Design of Adversarially Robust CNNsViability: 3.0
- Builds On ThisDetection of Adversarial Attacks in Robotic PerceptionViability: 2.0
- Prior WorkrSDNet: Unified Robust Neural Learning against Label Noise and Adversarial AttacksViability: 4.0
- Higher ViabilityAdversarial-Robust Multivariate Time-Series Anomaly Detection via Joint Information RetentionViability: 7.0
- Higher ViabilityRangeAD: Fast On-Model Anomaly DetectionViability: 7.0
Related Resources
- anomaly detection(glossary)
- How can geospatial AI be used for real-time anomaly detection in satellite imagery?(question)
- How can AI be used for real-time anomaly detection in aircraft systems?(question)
- How can Vision Transformers with semi-supervised training be used for anomaly detection in industrial settings?(question)