Timezone: »

On the Robustness of Deep Clustering Models: Adversarial Attacks and Defenses
Anshuman Chhabra · Ashwin Sekhari · Prasant Mohapatra

Thu Dec 01 09:00 AM -- 11:00 AM (PST) @ Hall J #205

Clustering models constitute a class of unsupervised machine learning methods which are used in a number of application pipelines, and play a vital role in modern data science. With recent advancements in deep learning-- deep clustering models have emerged as the current state-of-the-art over traditional clustering approaches, especially for high-dimensional image datasets. While traditional clustering approaches have been analyzed from a robustness perspective, no prior work has investigated adversarial attacks and robustness for deep clustering models in a principled manner. To bridge this gap, we propose a blackbox attack using Generative Adversarial Networks (GANs) where the adversary does not know which deep clustering model is being used, but can query it for outputs. We analyze our attack against multiple state-of-the-art deep clustering models and real-world datasets, and find that it is highly successful. We then employ some natural unsupervised defense approaches, but find that these are unable to mitigate our attack. Finally, we attack Face++, a production-level face clustering API service, and find that we can significantly reduce its performance as well. Through this work, we thus aim to motivate the need for truly robust deep clustering models.

Author Information

Anshuman Chhabra (University of California, Davis)
Anshuman Chhabra

Anshuman Chhabra is a Ph.D candidate at the University of California, Davis being advised by Prof. Prasant Mohapatra. Prior to that, he completed his B.Eng in Electronics and Communication Engineering from the University of Delhi, India. His research seeks to improve Machine Learning (ML) models and facilitate their adoption into society by analyzing model robustness along two dimensions: adversarial robustness (adversarial attacks/defenses against models) and social robustness (fair machine learning). His other research interests include designing Machine Learning and Reinforcement Learning based debiasing interventions for social media platforms such as YouTube and Twitter. He received the UC Davis Graduate Student Fellowship in 2018, and has held research positions at ESnet, Lawrence Berkeley National Laboratory, USA (2017), the Max Planck Institute for Software Systems, Germany (2020), and the University of Amsterdam, Netherlands (2022).

Ashwin Sekhari (University of California, Davis)
Prasant Mohapatra (University of California, Davis)

More from the Same Authors