Summary of Multi-normal Prototypes Learning For Weakly Supervised Anomaly Detection, by Zhijin Dong et al.
Multi-Normal Prototypes Learning for Weakly Supervised Anomaly Detection
by Zhijin Dong, Hongzhi Liu, Boyuan Ren, Weimin Xiong, Zhonghai Wu
First submitted to arxiv on: 23 Aug 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG)
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 novel framework for anomaly detection that addresses two common assumptions in existing methods: that normal samples cluster around a single central prototype, and that all unlabeled samples are normal. The proposed approach learns multiple normal prototypes using deep embedding clustering and contrastive learning, allowing it to efficiently work with limited labeled anomalies. Additionally, the method estimates the likelihood of each unlabeled sample being normal during training, enabling more efficient data encoding and normal prototype representation. Experimental results on various datasets demonstrate the superior performance of the proposed method compared to state-of-the-art approaches. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper introduces a new way to detect anomalies in different fields. It challenges common assumptions made by existing methods about how normal samples are grouped and whether all unknown samples are normal. The new approach learns many normal patterns using deep learning techniques, making it better at finding unusual data with limited information. This is useful for real-world applications. |
Keywords
» Artificial intelligence » Anomaly detection » Clustering » Deep learning » Embedding » Likelihood
