Review helps learn better: Temporal Supervised Knowledge Distillation

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wang, Dongwei, Han, Zhi, Wang, Yanmei, Chen, Xiai, Liu, Baichen, Tang, Yandong
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909606514524160
author Wang, Dongwei
Han, Zhi
Wang, Yanmei
Chen, Xiai
Liu, Baichen
Tang, Yandong
author_facet Wang, Dongwei
Han, Zhi
Wang, Yanmei
Chen, Xiai
Liu, Baichen
Tang, Yandong
contents Reviewing plays an important role when learning knowledge. The knowledge acquisition at a certain time point may be strongly inspired with the help of previous experience. Thus the knowledge growing procedure should show strong relationship along the temporal dimension. In our research, we find that during the network training, the evolution of feature map follows temporal sequence property. A proper temporal supervision may further improve the network training performance. Inspired by this observation, we propose Temporal Supervised Knowledge Distillation (TSKD). Specifically, we extract the spatiotemporal features in the different training phases of student by convolutional Long Short-term memory network (Conv-LSTM). Then, we train the student net through a dynamic target, rather than static teacher network features. This process realizes the refinement of old knowledge in student network, and utilizes it to assist current learning. Extensive experiments verify the effectiveness and advantages of our method over existing knowledge distillation methods, including various network architectures and different tasks (image classification and object detection) .
format Preprint
id arxiv_https___arxiv_org_abs_2307_00811
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Review helps learn better: Temporal Supervised Knowledge Distillation
Wang, Dongwei
Han, Zhi
Wang, Yanmei
Chen, Xiai
Liu, Baichen
Tang, Yandong
Computer Vision and Pattern Recognition
Artificial Intelligence
Reviewing plays an important role when learning knowledge. The knowledge acquisition at a certain time point may be strongly inspired with the help of previous experience. Thus the knowledge growing procedure should show strong relationship along the temporal dimension. In our research, we find that during the network training, the evolution of feature map follows temporal sequence property. A proper temporal supervision may further improve the network training performance. Inspired by this observation, we propose Temporal Supervised Knowledge Distillation (TSKD). Specifically, we extract the spatiotemporal features in the different training phases of student by convolutional Long Short-term memory network (Conv-LSTM). Then, we train the student net through a dynamic target, rather than static teacher network features. This process realizes the refinement of old knowledge in student network, and utilizes it to assist current learning. Extensive experiments verify the effectiveness and advantages of our method over existing knowledge distillation methods, including various network architectures and different tasks (image classification and object detection) .
title Review helps learn better: Temporal Supervised Knowledge Distillation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2307.00811