L2T-DLN: Learning to Teach with Dynamic Loss Network
This work addresses the challenge of adaptive teaching in machine learning, offering incremental improvements over existing methods by incorporating teacher experience and loss states.
The paper tackles the problem of improving student model training by introducing a teacher model with memory and dynamic loss networks, which uses past experience and loss states to adaptively adjust loss functions. The approach enhances student learning and improves performance across classification, object detection, and semantic segmentation tasks.
With the concept of teaching being introduced to the machine learning community, a teacher model start using dynamic loss functions to teach the training of a student model. The dynamic intends to set adaptive loss functions to different phases of student model learning. In existing works, the teacher model 1) merely determines the loss function based on the present states of the student model, i.e., disregards the experience of the teacher; 2) only utilizes the states of the student model, e.g., training iteration number and loss/accuracy from training/validation sets, while ignoring the states of the loss function. In this paper, we first formulate the loss adjustment as a temporal task by designing a teacher model with memory units, and, therefore, enables the student learning to be guided by the experience of the teacher model. Then, with a dynamic loss network, we can additionally use the states of the loss to assist the teacher learning in enhancing the interactions between the teacher and the student model. Extensive experiments demonstrate our approach can enhance student learning and improve the performance of various deep models on real-world tasks, including classification, objective detection, and semantic segmentation scenarios.