paper-with-me

Papers

Revisiting Architecture-aware Knowledge Distillation: Smaller Models and Faster Search

2022-06-27 · Taehyeon Kim, Heesoo Myeong, Se-Young Yun

Knowledge Distillation (KD) has recently emerged as a popular method for compressing neural networks. In recent studies, generalized distillation methods that find parameters and architectures of student models at the same time have been proposed. Still, this search method requires a lot of computation to search for architectures and has the disadvantage of considering only convolutional blocks in their search space. This paper introduces a new algorithm, coined as Trust Region Aware architecture search to Distill knowledge Effectively (TRADE), that rapidly finds effective student architectures from several state-of-the-art architectures using trust region Bayesian optimization approach. Experimental results show our proposed TRADE algorithm consistently outperforms both the conventional NAS approach and pre-defined architecture under KD training.

📄 PDF Abstract BibTeX arXiv:2206.13130

Code (0)

등록된 구현이 없습니다.

Tasks

Bayesian OptimizationKnowledge Distillation

Methods 이 논문이 사용한 방법론

AWARE We propose to theoretically and empirically examine the effect of incorporating weighting schemes into walk-aggregating GNNs. To this end, we propose a simple, interpretable, and…

Similar Papers 제목 키워드 기반

Revisiting Cross-Architecture Distillation: Adaptive Dual-Teacher Transfer for Lightweight Video Models

2025-11-12 · Ying Peng, Hongsen Ye, Changxin Huang, Xiping Hu 외 arxiv

Vision Transformers (ViTs) have achieved strong performance in video action recognition, but their high computational cost limits their practicality. Lightweight CNNs are more efficient but suffer from accuracy gaps. Cro…

Knowledge DistillationAction Recognition

Revisiting Data-Free Knowledge Distillation with Poisoned Teachers

2023-06-04 · Junyuan Hong, Yi Zeng, Shuyang Yu, Lingjuan Lyu 외

Data-free knowledge distillation (KD) helps transfer knowledge from a pre-trained model (known as the teacher model) to a smaller model (known as the student model) without access to the original training data used for t…

Backdoor Defense for Data-Free Distillation with Poisoned TeachersData-free Knowledge DistillationKnowledge Distillation

Revisiting Knowledge Distillation for Autoregressive Language Models

2024-02-19 · Qihuang Zhong, Liang Ding, Li Shen, Juhua Liu 외

Knowledge distillation (KD) is a common approach to compress a teacher model to reduce its inference cost and memory footprint, by training a smaller student model. However, in the context of autoregressive language mode…

Knowledge Distillation

Revisiting Self-Distillation

2022-06-17 · Minh Pham, Minsu Cho, Ameya Joshi, Chinmay Hegde

Knowledge distillation is the procedure of transferring "knowledge" from a large model (the teacher) to a more compact one (the student), often being used in the context of model compression. When both models have the sa…

Knowledge DistillationModel Compression

Towards Scalable & Efficient Interaction-Aware Planning in Autonomous Vehicles using Knowledge Distillation

2024-04-02 · Piyush Gupta, David Isele, Sangjae Bae

Real-world driving involves intricate interactions among vehicles navigating through dense traffic scenarios. Recent research focuses on enhancing the interaction awareness of autonomous vehicles to leverage these intera…

Autonomous VehiclesDecision MakingKnowledge DistillationModel Predictive Control+1