Summary of Unlearnable Examples For Time Series, by Yujing Jiang et al.
Unlearnable Examples For Time Series
by Yujing Jiang, Xingjun Ma, Sarah Monazam Erfani, James Bailey
First submitted to arxiv on: 3 Feb 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 This paper proposes a new method for generating unlearnable examples (UEs) to protect time series data from being exploited by deep learning models without authorization. The authors introduce a novel form of error-minimizing noise that can be selectively applied to specific segments of time series, making them unrecognizable to DNNs while remaining imperceptible to humans. The proposed method is tested on various datasets and demonstrated to be effective in both classification and generation tasks. This work contributes to the development of secure machine learning systems by protecting time series data from unauthorized training. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine you have a secret recipe, and someone wants to steal it without permission. To keep your recipe safe, this paper shows how to create “decoy” pieces of data that can’t be used by computers (like deep learning models) without authorization. The researchers developed a new way to add special noise to time series data (like stock prices or weather forecasts) that makes it unlearnable for machines while still looking normal to humans. They tested this method on many different datasets and showed it works well in both predicting what will happen next and generating new data. This helps make machine learning systems more secure and trustworthy. |
Keywords
* Artificial intelligence * Classification * Deep learning * Machine learning * Time series
