The ART of Transfer Learning: An Adaptive and Robust Pipeline
This work addresses a key bottleneck in transfer learning for machine learning practitioners by preventing negative transfer, though it appears incremental as it builds on existing methods with a flexible pipeline.
The authors tackled the problem of negative transfer in transfer learning by proposing the Adaptive Robust Transfer Learning (ART) pipeline, which achieved adaptive transfer with theoretical guarantees and demonstrated promising performance across regression, classification, and sparse learning tasks.
Transfer learning is an essential tool for improving the performance of primary tasks by leveraging information from auxiliary data resources. In this work, we propose Adaptive Robust Transfer Learning (ART), a flexible pipeline of performing transfer learning with generic machine learning algorithms. We establish the non-asymptotic learning theory of ART, providing a provable theoretical guarantee for achieving adaptive transfer while preventing negative transfer. Additionally, we introduce an ART-integrated-aggregating machine that produces a single final model when multiple candidate algorithms are considered. We demonstrate the promising performance of ART through extensive empirical studies on regression, classification, and sparse learning. We further present a real-data analysis for a mortality study.