Lab member paper accepted at IEEE 2022

• Post by: Bahareh Arghavani
• February 17, 2023
• Comments off

Congratulations on the acceptance of your paper titled “Adversarial Label-Poisoning Attacks and Defense for General Multi-Class Models Based on Synthetic Reduced Nearest Neighbor” This is a significant achievement that reflects your hard work, dedication, and expertise in the field of machine learning and artificial intelligence.

The paper is available on IEEE.

Your research has great potential to make a real impact on the development of more robust and reliable AI systems, which is an essential task in today’s rapidly evolving technological landscape. Your approach to using the Theory of Mind framework to mitigate adversarial communication at test time is innovative and promising, and we look forward to seeing the results of further research and development in this area.

Abstract :

Machine learning models are vulnerable to data poisoning attacks whose purpose is to undermine the model’s integrity. However, the current literature on data poisoning attacks mainly focuses on ad hoc techniques that are generally limited to either binary classifiers or to gradient-based algorithms. To address these limitations, we propose a novel model-free label-flipping attack based on the multi-modality of the data, in which the adversary targets the clusters of classes while constrained by a label-flipping budget. The complexity of our proposed attack algorithm is linear in time over the size of the dataset. Also, the proposed attack can increase the error up to two times for the same attack budget. Second, a novel defense technique is proposed based on the Synthetic Reduced Nearest Neighbor model. The defense technique can detect and exclude flipped samples on the fly during the training procedure. Our empirical analysis demonstrates that (i) the proposed attack technique can deteriorate the accuracy of several models drastically, and (ii) under the proposed attack, the proposed defense technique significantly outperforms other conventional machine learning models in recovering the accuracy of the targeted model.

One-sentence Summary: We propose a novel fast decision tree induction algorithm, outperforming the state-of-the-art by several orders of magnitude.

Author(s):

Pooya Tavallali; Vahid Behzadan; Azar Alizadeh; Aditya Ranganath; Mukesh Singhal