Outlier detection in deep neural networks
By using t-path feature combination and threshold calculation in deep neural networks, the unpredictability of DNN models on untrained data is solved, achieving efficient detection and accurate labeling of outliers and reducing mislabeling of normal images.
CN116034377BActive Publication Date: 2026-06-23INTERNATIONAL BUSINESS MACHINE CORPORATION
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- INTERNATIONAL BUSINESS MACHINE CORPORATION
- Filing Date
- 2021-09-20
- Publication Date
- 2026-06-23
Smart Images
Figure CN116034377B_ABST
Abstract
Anomaly detection using a deep neural network (DNN) includes running a trained DNN model on a received input item. A first feature vector is extracted from the input item and quantized to discrete values. A first number of special t-way feature combinations is computed in the input item and compared to a computed threshold. Based on the comparison, the input item is flagged as an anomaly and an alert is generated notifying of the flagged input item.
Need to check novelty before this filing date? Find Prior Art