Summary of Enhancing Robustness Of Data-driven Shm Models: Adversarial Training with Circle Loss, by Xiangli Yang et al.
Enhancing robustness of data-driven SHM models: adversarial training with circle loss
by Xiangli Yang, Xijie Deng, Hanwei Zhang, Yang Zou, Jianxi Yang
First submitted to arxiv on: 20 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 proposed paper aims to address the vulnerability of machine learning-based structural health monitoring (SHM) systems to adversarial attacks. In SHM, machine learning models are used to analyze data from sensors and predict potential failures or damages. However, these models can be manipulated by introducing small changes in input data, leading to different outputs. To mitigate this issue, the paper proposes an adversarial training method that uses circle loss to optimize the distance between features in training data, keeping examples away from the decision boundary. The proposed approach demonstrates substantial improvements in model robustness, surpassing existing defense mechanisms. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The research aims to make structural health monitoring systems more reliable by protecting them against fake or manipulated data. This is important because these systems help keep buildings and bridges safe. Right now, there are models that can be tricked into making wrong predictions if the input data is changed slightly. The proposed solution tries to fix this problem by training the model in a special way that makes it more robust to small changes in data. |
Keywords
» Artificial intelligence » Machine learning
