Handling Class Imbalance in Black-Box Unsupervised Domain Adaptation with Synthetic Minority Over-Sampling

Keyphrases

• Synthetic Minority Oversampling Technique (SMOTE) 100%
• Unsupervised Domain Adaptation 100%
• Class Imbalance 100%
• Minority Class 50%
• Majority Class 25%
• Pseudo Label 25%
• Rebalance 25%
• Low Sampling Rate 25%
• Synthetic Samples 25%
• Data Security 25%
• Target Domain 25%

Computer Science

• Unsupervised Domain Adaptation 100%
• Class Imbalance 100%
• Minority Class 66%
• Data Security 33%
• Majority Class 33%
• Low Sampling Rate 33%