Active Learning Comparison Hub
3 papers - avg viability 7.0
Top Papers
- Data Agent: Learning to Select Data via End-to-End Dynamic Optimization(7.0)
Data Agent accelerates model training by dynamically selecting informative samples, reducing training costs by up to 50% while maintaining performance.
- Revisiting Unknowns: Towards Effective and Efficient Open-Set Active Learning(7.0)
E2OAL is a unified, detector-free framework for open-set active learning that improves accuracy and efficiency by leveraging labeled unknowns, making it suitable for real-world applications.
- Adaptive Active Learning for Regression via Reinforcement Learning(7.0)
WiGS leverages reinforcement learning to optimize active learning for regression, enhancing labeling efficiency and accuracy.