Summary of Manipulating Feature Visualizations with Gradient Slingshots, by Dilyara Bareeva et al.
Manipulating Feature Visualizations with Gradient Slingshots
by Dilyara Bareeva, Marina M.-C. Höhne, Alexander Warnecke, Lukas Pirch, Klaus-Robert Müller, Konrad Rieck, Kirill Bykov
First submitted to arxiv on: 11 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
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 The paper investigates the vulnerability of Feature Visualization (FV), a popular technique used to explain Deep Neural Networks’ (DNNs) learned concepts. The authors propose a novel approach that manipulates FV without affecting the DNN’s decision-making process, and demonstrate its effectiveness on various neural network models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper explores how to trick Feature Visualization (FV), which shows what a deep learning model is looking at when it makes decisions. The researchers found a way to make FV show something else instead, without changing the way the model works. They tested this new approach with different types of neural networks and showed that it can hide what’s really going on inside these models. |
Keywords
* Artificial intelligence * Deep learning * Neural network
