paper-with-me

홈 › Papers

Generalized Knowledge Distillation via Relationship Matching

2022-05-04 · Han-Jia Ye, Su Lu, De-Chuan Zhan

The knowledge of a well-trained deep neural network (a.k.a. the "teacher") is valuable for learning similar tasks. Knowledge distillation extracts knowledge from the teacher and integrates it with the target model (a.k.a. the "student"), which expands the student's knowledge and improves its learning efficacy. Instead of enforcing the teacher to work on the same task as the student, we borrow the knowledge from a teacher trained from a general label space -- in this "Generalized Knowledge Distillation (GKD)", the classes of the teacher and the student may be the same, completely different, or partially overlapped. We claim that the comparison ability between instances acts as an essential factor threading knowledge across tasks, and propose the RElationship FacIlitated Local cLassifiEr Distillation (REFILLED) approach, which decouples the GKD flow of the embedding and the top-layer classifier. In particular, different from reconciling the instance-label confidence between models, REFILLED requires the teacher to reweight the hard tuples pushed forward by the student and then matches the similarity comparison levels between instances. An embedding-induced classifier based on the teacher model supervises the student's classification confidence and adaptively emphasizes the most related supervision from the teacher. REFILLED demonstrates strong discriminative ability when the classes of the teacher vary from the same to a fully non-overlapped set w.r.t. the student. It also achieves state-of-the-art performance on standard knowledge distillation, one-step incremental learning, and few-shot learning tasks.

📄 PDF Abstract BibTeX arXiv:2205.01915

Code (1)

njulus/gkd 공식 구현 pytorch

Tasks

Few-Shot LearningIncremental LearningKnowledge Distillation

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

Generative Dataset Distillation Based on Self-knowledge Distillation

2025-01-08 · Longzhen Li, Guang Li, Ren Togo, Keisuke Maeda 외

Dataset distillation is an effective technique for reducing the cost and complexity of model training while maintaining performance by compressing large datasets into smaller, more efficient versions. In this paper, we p…

Dataset DistillationKnowledge DistillationSelf-Knowledge Distillation

G-DetKD: Towards General Distillation Framework for Object Detectors via Contrastive and Semantic-guided Feature Imitation

2021-08-17 · ICCV 2021 10 · Lewei Yao, Renjie Pi, Hang Xu, Wei zhang 외

In this paper, we investigate the knowledge distillation (KD) strategy for object detection and propose an effective framework applicable to both homogeneous and heterogeneous student-teacher pairs. The conventional feat…

Knowledge Distillationobject-detectionObject Detection

Lightweight SAR Ship Detection via Contrastive Distillation

2026-05-27 · Surendar Devasundaram, Banafsheh Saber Latibari, Abhijit Mahalanobis arxiv

Deep convolutional and transformer-based detectors achieve strong performance for SAR ship detection but are often computationally prohibitive for real-time or onboard deployment. Lightweight models offer improved effici…

Knowledge Distillation

Distilling Cross-Task Knowledge via Relationship Matching

2020-06-01 · CVPR 2020 6 · Han-Jia Ye, Su Lu, De-Chuan Zhan

The discriminative knowledge from a high-capacity deep neural network (a.k.a. the "teacher") could be distilled to facilitate the learning efficacy of a shallow counterpart (a.k.a. the "student"). This paper deals with a…

Knowledge Distillation

Rethinking Decoupled Knowledge Distillation: A Predictive Distribution Perspective

2025-12-04 · Bowen Zheng, Ran Cheng arxiv

In the history of knowledge distillation, the focus has once shifted over time from logit-based to feature-based approaches. However, this transition has been revisited with the advent of Decoupled Knowledge Distillation…

Knowledge Distillation