Training Description
Key Takeaways
- Focuses on security and privacy challenges in machine learning (ML) systems
- Explores adversarial learning, robust algorithms, and privacy-preserving techniques
- Addresses risks related to Artificial General Intelligence (AGI) and mis/disinformation
- Targets academic researchers, industry professionals, and practitioners in ML/AI security
- Emphasizes bridging theoretical advances with real-world secure ML applications
The 8th International Workshop on Security in Machine Learning and Its Applications (SIMLA) is an academic event dedicated to advancing the understanding of security and privacy in machine learning. Held alongside the ACNS2026 conference, this workshop brings together experts to discuss the latest challenges and solutions in safeguarding ML systems.
Workshop Overview
SIMLA provides a specialized forum for the exchange of ideas on the vulnerabilities and risks associated with the widespread adoption of machine learning technologies. The workshop highlights the importance of adversarial machine learning, robust algorithm design, and privacy-preserving methods. As ML systems become integral to various industries, the need for secure and trustworthy deployment grows increasingly critical.
Participants engage in discussions on emerging threats, such as adversarial attacks and the manipulation of AI models. The event also addresses the challenges of content provenance and the detection of mis/disinformation, especially in the context of advanced AI and AGI systems.
Main Topics and Themes
The workshop covers a range of topics, including adversarial learning, robust evaluation of ML algorithms, and privacy-preserving techniques. Attendees explore methods for secure deployment of ML systems and strategies to mitigate the misuse of AI agents. The event also examines the implications of AGI, focusing on the trustworthy and safe integration of advanced AI technologies.
Recurring themes include the security and privacy implications of ML systems, the development of robust and resilient models, and the application of theoretical advances to practical, real-world scenarios.
Audience and Experience
SIMLA is designed for academic researchers, industry practitioners, and professionals in cybersecurity, machine learning, and AI. The workshop fosters a collaborative environment for sharing knowledge, networking, and recognizing outstanding contributions, such as the best paper award sponsored by Springer.
Held in-person in parallel with the ACNS2026 main conference, the event offers a technical and interactive experience, supporting education, thought leadership, and community building within the field of secure and privacy-preserving machine learning.