Amalgamating Filtered Knowledge: Learning Task-customized Student from Multi-task Teachers
This work addresses the challenge of efficiently adapting pre-trained models to new, customized tasks in computer vision, though it is incremental in its approach.
The paper tackles the problem of learning a customized student network for specific tasks using multiple pre-trained teacher models without human-labeled annotations, achieving results that outperform the teachers on those tasks.
Many well-trained Convolutional Neural Network(CNN) models have now been released online by developers for the sake of effortless reproducing. In this paper, we treat such pre-trained networks as teachers and explore how to learn a target student network for customized tasks, using multiple teachers that handle different tasks. We assume no human-labelled annotations are available, and each teacher model can be either single- or multi-task network, where the former is a degenerated case of the latter. The student model, depending on the customized tasks, learns the related knowledge filtered from the multiple teachers, and eventually masters the complete or a subset of expertise from all teachers. To this end, we adopt a layer-wise training strategy, which entangles the student's network block to be learned with the corresponding teachers. As demonstrated on several benchmarks, the learned student network achieves very promising results, even outperforming the teachers on the customized tasks.