Data Efficient
Data-efficient machine learning focuses on developing algorithms and techniques that achieve high performance with significantly less training data than traditional methods. Current research emphasizes strategies like data pruning, active learning (selecting the most informative samples), and the use of pre-trained models for transfer learning across various architectures including diffusion models, transformers, and neural networks. This field is crucial for addressing limitations in data availability, computational resources, and privacy concerns, impacting diverse applications from materials science and robotics to reinforcement learning and natural language processing.
98papers
Papers - Page 4
October 23, 2023
October 16, 2023
September 18, 2023
August 8, 2023
June 28, 2023