Anomalous Instance Detection in Deep Learning: A Survey

By Saikiran Bulusu, Bhavya Kailkhura, Bo Li, Pramod K. Varshney, Dawn Song. 2020

48 citations recorded by the source corpus.

Deep Learning (DL) is vulnerable to out-of-distribution and adversarial examples resulting in incorrect outputs. To make DL more robust, several posthoc anomaly detection techniques to detect (and discard) these anomalous samples have been proposed in the recent past. This survey tries to provide a structured and comprehensive overview of the research on anomaly detection for DL based applications. We provide a taxonomy for existing techniques based on their underlying assumptions and adopted approaches. We discuss various techniques in each of the categories and provide the relative strengths and weaknesses of the approaches. Our goal in this survey is to provide an easier yet better understanding of the techniques belonging to different categories in which research has been done on this topic. Finally, we highlight the unsolved research challenges while applying anomaly detection techniques in DL systems and present some high-impact future research directions.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Gyuhak Kim
2. Changnan Xiao
3. Sepideh Esmaeilpour
4. Shuhan Yuan
5. Hui Xiong
6. Jia Wu
7. Quan Z. Sheng
8. Chuan Zhou

In Signal Processing

1. Sepideh Esmaeilpour
2. Hui Xiong
3. Gyuhak Kim
4. Haixuan Guo
5. Changnan Xiao
6. Shuhan Yuan
7. Bing Liu
8. Xintao Wu

Citations in the corpus

• A Theoretical Study on Solving Continual Learning (2022, neutral): “Excellent surveys of OOD detection include [60, 61].”
• Continual Learning Based on OOD Detection and Task Masking (2022, methodology): “OOD detection is stated as follows [4]: Given a training set of n classes, called the in-distribution (IND) training data, we want to build a model that can assign correct classes to IND test data and reject or detect OOD test data that do not belong to any of the n IND training classes.”
• A Comprehensive Survey on Graph Anomaly Detection With Deep Learning (2021, background): “Specifically, [42], [43] and [41] are representative surveys on the generalized anomaly detection techniques.”
• LogBERT: Log Anomaly Detection via BERT (2021, background): “The intrinsic characteristics of the boundary are dynamically evolving, which poses challenges for both machine learning-based and deep learning-based algorithms [17].”