Summary of Evaluating Deep Learning Models For Fault Diagnosis Of a Rotating Machinery with Epistemic and Aleatoric Uncertainty, by Reza Jalayer et al.
Evaluating deep learning models for fault diagnosis of a rotating machinery with epistemic and aleatoric uncertainty
by Reza Jalayer, Masoud Jalayer, Andrea Mor, Carlotta Orsenigo, Carlo Vercellis
First submitted to arxiv on: 25 Dec 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 presents a comprehensive comparative study of state-of-the-art uncertainty-aware deep learning (DL) architectures for fault diagnosis in rotating machinery. The selected architectures include sampling by dropout, Bayesian neural networks, and deep ensembles. To distinguish between in-distribution and out-of-distribution (OOD) data, two uncertainty thresholds are applied. Empirical findings reveal that all DL models can effectively detect OOD data in the presence of epistemic uncertainty, with deep ensemble models showing superior performance. In the presence of aleatoric uncertainty, the noise level plays a crucial role, and low noise levels hinder model performance. However, deep ensemble models exhibit milder degradation in performance, making them the preferred choice due to their shorter inference time. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper compares different types of artificial intelligence (AI) models for detecting problems with machines that spin around. These AI models are special because they can understand when they don’t have enough information or if there’s noise in the data. The study shows that one type of AI model, called deep ensemble, is best at detecting unknown problems and performs well even when there’s some noise. This is important because it helps us build better machines that can detect issues before they cause damage. |
Keywords
» Artificial intelligence » Deep learning » Dropout » Inference
