Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 21 hours
Course Outline
Intro to TinyML Security
- Challenges of security in resource-limited ML systems
- Threat modeling for TinyML implementations
- Risk classifications for embedded AI use cases
Edge AI Data Privacy
- Privacy implications for processing data on-device
- Reducing data exposure and transmission
- Strategies for distributed data management
Countering Adversarial Attacks on TinyML
- Threats from model evasion and data poisoning
- Manipulating inputs on embedded sensors
- Evaluating vulnerabilities in constrained settings
Hardening Embedded ML Security
- Protection layers for firmware and hardware
- Secure boot and access control protocols
- Best practices for protecting inference workflows
Privacy-Focused TinyML Methods
- Privacy considerations in quantization and model design
- On-device data anonymization techniques
- Lightweight cryptography and secure computation
Secure Deployment and Upkeep
- Secure provisioning for TinyML devices
- OTA updates and patch management strategies
- Edge-level monitoring and incident response
Testing and Validating Secure TinyML
- Frameworks for security and privacy testing
- Simulating realistic attack vectors
- Compliance and validation considerations
Case Studies and Practical Applications
- Security incidents in edge AI environments
- Building robust TinyML architectures
- Assessing the balance between performance and security
Conclusion and Future Directions
Requirements
- Familiarity with embedded system architectures
- Background in machine learning workflows
- Foundational knowledge of cybersecurity
Target Audience
- Security analysts
- AI developers
- Embedded systems engineers
Testimonials (1)
The exercises
