2019/08/29 by Linchao Zhu, Sercan Ö. Arık, Zhu, Linchao +6
Computer Science · #Adaptation (eye) #Artificial intelligence #Computer science #Domain Adaptation and Few-Shot Learning #Machine Learning and Data Classification #Machine learning #Metric (unit) #Multimodal Machine Learning Applications #Reinforcement learning #Selection (genetic algorithm) #Set (abstract data type) #Transfer of learning #cs.AI #cs.LG
paper · pdf · doi:10.48550/arxiv.1908.11406
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2019/08/29 · arxiv created 2020/07/16 · arxiv updated 2020/07/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/08
We propose a novel adaptive transfer learning framework, learning to transfer learn (L2TL), to improve performance on a target dataset by careful extraction of the related information from a source dataset. Our framework considers cooperative optimization of shared weights between models for source and target tasks, and adjusts the constituent loss weights adaptively. The adaptation of the weights is based on a reinforcement learning (RL) selection policy, guided with a performance metric on the target validation set. We demonstrate that L2TL outperforms fine-tuning baselines and other adaptive transfer learning methods on eight datasets. In the regimes of small-scale target datasets and significant label mismatch between source and target datasets, L2TL shows particularly large benefits.