SARA 2020: First Workshop on Secure and Resilient Autonomy
January 26, 2020
First Workshop on Secure and Resilient Autonomy (SARA) at MLSys 2020
March 4, 2020
Submissions Due: January 26, 2020
This workshop will bring classical system architecture and design experts and AI/ML algorithmic experts together in one forum. The goal is to brainstorm about challenges in designing secure and resilient AI-centric systems in general, but with a special focus on autonomous systems (such as self-driving cars and industrial robots) – where safety and security are of paramount value. The knowledge and expertise of classical mainframe and server architects who are experts in designing ultra-reliable and secure systems will be blended with domain experts in AI; particularly those with an established expertise in developing reliable and secure AI algorithms. The organizers of this workshop largely represent the classical system architects with expertise in building robust and energy efficient systems. The program will be a blend of talks selected from submitted abstracts and invited speakers. The latter will largely feature experts in core AI algorithms, especially those focused on adversarial robustness, few-shot learning, immunity against catastrophic forgetfulness, etc.
Call for presentation abstracts:
This first workshop on SARA (Secure and Resilient Autonomy), at MLSys 2020 will primarily focus on the security and reliability aspects of AI-centric systems. Topics of interest include (but are not limited to):
• Resilience and security considerations in emerging new domains such as autonomous vehicles, cognitive IoT swarms, as well as existing server-class and embedded architectures for the AI and machine learning applications.
• Anomaly detection in AI-centric systems.
• Error and threat models; associated vulnerability assessment metrics.
• Design of secure and robust cloud-backed edge-architectures for machine learning applications.
• Algorithmic techniques to improve training and inference in the presence of errors.
• Hardware-software co-design of resilient machine learning systems.
• Characterization of hardware and system-level vulnerabilities resulting in unplanned failures and/or adversarial attacks.
• End-to-end resilience evaluation of real machine learning systems.
• Resilient design of distributed swarm-based architectures for machine learning.
• Lifelong learning, few-shot learning and mitigation of catastrophic forgetfulness.
• Energy efficiency and endurance in mobile and embedded AI architectures.
Researchers in this field are encouraged to submit an extended abstract. We also encourage presentations showcasing prototype demonstrations and open source contributions.
• Submission of presentation abstracts: January 26th 2020.
• Notification of acceptance: February 7th 2020.
• Final 2-page (max) paper due (for inclusion in online proceedings): February 24th 2020.
• Workshop date: March 4th 2020.
• Workshop venue: Austin, TX (co-located with MLSys 2020).
• Pradip Bose, IBM T. J. Watson Research Center
• Nandhini Chandramoorthy, IBM T. J. Watson Research Center
• Augusto Vega, IBM T. J. Watson Research Center
• Karthik Swaminathan, IBM T. J. Watson Research Center
If you have questions regarding submission, please contact us: firstname.lastname@example.org