Summary of Adversarial Attacks and Dimensionality in Text Classifiers, by Nandish Chattopadhyay et al.
Adversarial Attacks and Dimensionality in Text Classifiers
by Nandish Chattopadhyay, Atreya Goswami, Anupam Chattopadhyay
First submitted to arxiv on: 3 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
Summary difficulty | Written by | Summary |
---|---|---|
High | Paper authors | High Difficulty Summary Read the original abstract here |
Medium | GrooveSquid.com (original content) | Medium Difficulty Summary In this paper, researchers investigate the effectiveness of adversarial attacks on machine learning algorithms in natural language processing (NLP) tasks. Specifically, they study text classification tasks and find that there is a strong correlation between the dimensionality of the model’s embeddings and the success of these attacks. They utilize this sensitivity to design an adversarial defense mechanism based on ensemble models with varying dimensionality, which is tested for its efficacy on multiple datasets. Additionally, the paper explores different distance metrics for measuring adversarial perturbations. |
Low | GrooveSquid.com (original content) | Low Difficulty Summary This study shows that machine learning algorithms are vulnerable to tiny but strategic changes in text data, making them unreliable for real-world use cases. The researchers discovered a link between the dimensionality of word embeddings and the success of these attacks. They then developed an anti-attack strategy using multiple models with different dimensions. This was tested on various datasets to see how well it worked. |
Keywords
» Artificial intelligence » Machine learning » Natural language processing » Nlp » Text classification