CDC-Calibrated Deep Clustering Framework: 10倍の期待されるキャリブレーションエラー改善
State-of-the-art deep clustering methods: 約10倍の期待されるキャリブレーションエラー改善
引用
"Our model not only surpasses state-of-the-art deep clustering methods by approximately 10 times in terms of expected calibration error but also significantly outperforms them in terms of clustering accuracy."
"The proposed calibrated deep clustering framework tackles the overconfidence problem and significantly improves both reliability and accuracy."