Abstract

In the ever-evolving landscape of artificial intelligence (AI), we find ourselves at a pivotal juncture. AI is no longer confined to theoretical musings or laboratory experiments; it has taken flight, soaring on the wings of innovation and practical application. As we delve into the heart of this transformation, we witness the emergence of AI copilots—intelligent companions that augment human capabilities and navigate complex domains alongside us. In this talk, we embark on a journey through the realms of applied computer vision research at Microsoft. Our mission? To empower everyone on the planet to achieve more using the best technologies. We explore how computer vision, a cornerstone of AI, transcends mere algorithms and pixels. It becomes a conduit for real-world impact, touching lives, industries, and societies. Together, we ride the currents of innovation, steering toward a horizon where AI’s wings carry us to uncharted heights. Machine learning models are vulnerable to adversarial attacks that add imperceptible perturbations to the test data. However, in most real-world applications, the attacker only has limited knowledge about the victim model, making such kind of attacks difficult to succeed. The attack is even more challenging for targeted attacks, where the attacker wants to mislead the model into a specific prediction outcome.
In this talk, I will introduce several techniques for improving the targeted transfer abilities of adversarial attacks. I will also introduce defenses and countermeasures, including a novel adversarial training method.

Bio.

Shang-Tse Chen is an Assistant Professor at NTU CSIE. He works at the intersection of applied and theoretical machine learning, with a strong application focus on cybersecurity. His research has led to patented cyber threat detection technology with Symantec, open-sourced adversarial attack and defense tools with Intel, and deployed fire risk prediction system with the Atlanta Fire Rescue Department. His recent research interests include adversarial ML and various aspects of security, privacy, and fairness of ML models.

All faculty and students are welcome to join.