Using synthetic data and unsupervised learning methods for malware detection

使用合成数据和无监督学习方法进行恶意软件检测

• 批准号: 10099768
• 金额: $ 7.53万
• 依托单位: UNIVERSITY OF STRATHCLYDE
• 依托单位国家: 英国
• 项目类别: Collaborative R&D
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=10099768
• 关键词: Using; synthetic; data; unsupervised; learning

In today's AI-driven world, with great innovation comes great?vulnerability.?ECG.ai is the game-changer. We offer a?tailored?cybersecurity platform for AI systems, using synthetic data?to mimic?and defend against vulnerabilities.??We are?not just reactive, we are proactive, seamlessly integrating into?existing development processes and always staying steps ahead of threats.?As AI adoption surges, regulatory focus sharpens.?ECG.ai ensures AI?systems are not just smart, but secure and compliant.??ECG.ai is a cybersecurity platform that secures AI systems against adversarial threats. It provides a modular API that can be integrated into the existing development lifecycle to generate synthetic data that can?be used to increase defence capabilities of AI models and simulates edge cases and novel attacks to improve model resilience. Our platform allows organizations to bridge the gap between cybersecurity and ethical AI using synthetic data to understand new threats and harden the defensive capabilities of AI models.

在当今人工智能驱动的世界中，伟大的创新也伴随着巨大的脆弱性。ECG.ai 是游戏规则的改变者。我们为人工智能系统提供“量身定制”的网络安全平台，使用合成数据来模拟和防御漏洞。我们不仅是被动的，我们是主动的，无缝集成到现有的开发流程中，并始终保持领先地位威胁。随着人工智能采用的激增，监管焦点变得更加尖锐。ECG.ai 确保人工智能系统不仅智能，而且安全且合规。ECG.ai 是一个网络安全平台，可保护人工智能系统免受对抗性威胁。它提供了一个模块化的 API，可以集成到现有的开发生命周期中，生成合成数据，用于提高 AI 模型的防御能力，并模拟边缘情况和新颖的攻击，以提高模型的弹性。我们的平台允许组织使用合成数据来了解新威胁并强化人工智能模型的防御能力，从而弥合网络安全和道德人工智能之间的差距。