paper-with-me

홈 › Papers

It's All in the Head: Representation Knowledge Distillation through Classifier Sharing

2022-01-18 · Emanuel Ben-Baruch, Matan Karklinsky, Yossi Biton, Avi Ben-Cohen, Hussam Lawen, Nadav Zamir

Representation knowledge distillation aims at transferring rich information from one model to another. Common approaches for representation distillation mainly focus on the direct minimization of distance metrics between the models' embedding vectors. Such direct methods may be limited in transferring high-order dependencies embedded in the representation vectors, or in handling the capacity gap between the teacher and student models. Moreover, in standard knowledge distillation, the teacher is trained without awareness of the student's characteristics and capacity. In this paper, we explore two mechanisms for enhancing representation distillation using classifier sharing between the teacher and student. We first investigate a simple scheme where the teacher's classifier is connected to the student backbone, acting as an additional classification head. Then, we propose a student-aware mechanism that asks to tailor the teacher model to a student with limited capacity by training the teacher with a temporary student's head. We analyze and compare these two mechanisms and show their effectiveness on various datasets and tasks, including image classification, fine-grained classification, and face verification. In particular, we achieve state-of-the-art results for face verification on the IJB-C dataset for a MobileFaceNet model: TAR@(FAR=1e-5)=93.7\%. Code is available at https://github.com/Alibaba-MIIL/HeadSharingKD.

📄 PDF Abstract BibTeX arXiv:2201.06945

Code (1)

alibaba-miil/headsharingkd 공식 구현 pytorch

Tasks

AllClassificationFace Verificationimage-classificationImage ClassificationKnowledge DistillationTAR

Methods 이 논문이 사용한 방법론

Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…

Similar Papers 제목 키워드 기반

Distilling Knowledge via Intermediate Classifiers

2021-02-28 · Aryan Asadian, Amirali Salehi-Abari

The crux of knowledge distillation is to effectively train a resource-limited student model with the guide of a pre-trained larger teacher model. However, when there is a large difference between the model complexities o…

Knowledge DistillationTransfer Learning

Personalized Federated Learning through Global Knowledge Distillation and Local Head Adaptation

2026-09-15 · Polycarpo Souza Neto, José Mairton Barros da Silva Júnior, Charles Casimiro Cavalcante arxiv

Statistical heterogeneity limits federated learning when a single global classifier cannot represent client-specific label distributions. In this work, we propose Personalized Federated Knowledge Distillation with Head A…

Personalized Federated LearningKnowledge Distillation

Multi-head Knowledge Distillation for Model Compression

2020-12-05 · Huan Wang, Suhas Lohit, Michael Jones, Yun Fu

Several methods of knowledge distillation have been developed for neural network compression. While they all use the KL divergence loss to align the soft outputs of the student model more closely with that of the teacher…

image-classificationImage ClassificationKnowledge Distillationmodel+2

Pushing the Limits of Distillation-Based Continual Learning via Classifier-Proximal Lightweight Plugins

2025-12-03 · Zhiming Xu, Baile Xu, Jian Zhao, Furao Shen 외 arxiv

Continual learning requires models to learn continuously while preserving prior knowledge under evolving data streams. Distillation-based methods are appealing for retaining past knowledge in a shared single-model framew…

Continual Learning

Knowledge Distillation Meets Open-Set Semi-Supervised Learning

2022-05-13 · Jing Yang, Xiatian Zhu, Adrian Bulat, Brais Martinez 외

Existing knowledge distillation methods mostly focus on distillation of teacher's prediction and intermediate activation. However, the structured representation, which arguably is one of the most critical ingredients of …

Face RecognitionKnowledge Distillation