.png){kind=icon}
返回
Dynamic Correlation Learning and Regularization for Multi-Label Confidence Calibration
.png){kind=icon}
PRE
AI
DOI：10.1109/TIP.2024.3448248.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Modern visual recognition models often display overconfidence due to their reliance on complex deep neural networks and one-hot target supervision, resulting in unreliable confidence scores that necessitate calibration. While current confidence calibration techniques primarily address single-label scenarios, there is a lack of focus on more practical and generalizable multi-label contexts. This paper introduces the Multi-Label Confidence Calibration (MLCC) task, aiming to provide well-calibrated confidence scores in multi-label scenarios. Unlike single-label images, multi-label images contain multiple objects, leading to semantic confusion and further unreliability in confidence scores. Existing single-label calibration methods, based on label smoothing, fail to account for category correlations, which are crucial for addressing semantic confusion, thereby yielding sub-optimal performance. To overcome these limitations, we propose the Dynamic Correlation Learning and Regularization (DCLR) algorithm, which leverages multi-grained semantic correlations to better model semantic confusion for adaptive regularization. DCLR learns dynamic instance-level and prototype-level similarities specific to each category, using these to measure semantic correlations across different categories. With this understanding, we construct adaptive label vectors that assign higher values to categories with strong correlations, thereby facilitating more effective regularization. We establish an evaluation benchmark, re-implementing several advanced confidence calibration algorithms and applying them to leading multi-label recognition (MLR) models for fair comparison. Through extensive experiments, we demonstrate the superior performance of DCLR over existing methods in providing reliable confidence scores in multi-label scenarios.
Keyword:
Calibration
Correlation
Semantics
Predictive models
Adaptation models
Vectors
Task analysis
Multi-label image recognition
confidence calibration
over-confidence
trusted artificial intelligence
期刊
IF:
13.7
论文数:
1.0W
被引数:
8.4W
机构
引用论文
Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations视觉基因组: 使用众包密集图像注释连接语言和视觉
OAAI
OAAI分享收藏Dynamically Weighted Balanced Loss: Class Imbalanced Learning and Confidence Calibration of Deep Neural Networks动态加权平衡损失: 深度神经网络的类不平衡学习和置信度校准
PREAI
分享收藏Femtosecond Charge Transfer Dynamics of a Modified DNA Base: 2-Aminopurine in Complexes with Nucleotides.png)
ChemPhysChemOAAI
ChemPhysChemIF0
OAAI分享收藏
查看更多内容