paper-with-me

홈 › Papers

Towards Optimal Trade-offs in Knowledge Distillation for CNNs and Vision Transformers at the Edge

2024-06-25 · John Violos, Symeon Papadopoulos, Ioannis Kompatsiaris

This paper discusses four facets of the Knowledge Distillation (KD) process for Convolutional Neural Networks (CNNs) and Vision Transformer (ViT) architectures, particularly when executed on edge devices with constrained processing capabilities. First, we conduct a comparative analysis of the KD process between CNNs and ViT architectures, aiming to elucidate the feasibility and efficacy of employing different architectural configurations for the teacher and student, while assessing their performance and efficiency. Second, we explore the impact of varying the size of the student model on accuracy and inference speed, while maintaining a constant KD duration. Third, we examine the effects of employing higher resolution images on the accuracy, memory footprint and computational workload. Last, we examine the performance improvements obtained by fine-tuning the student model after KD to specific downstream tasks. Through empirical evaluations and analyses, this research provides AI practitioners with insights into optimal strategies for maximizing the effectiveness of the KD process on edge devices.

📄 PDF Abstract BibTeX arXiv:2407.12808

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge Distillation

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Residual Connection 설명 없음
Adam 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Similar Papers 제목 키워드 기반

Towards Higher Pareto Frontier in Multilingual Machine Translation

2023-05-25 · Yichong Huang, Xiaocheng Feng, Xinwei Geng, Baohang Li 외

Multilingual neural machine translation has witnessed remarkable progress in recent years. However, the long-tailed distribution of multilingual corpora poses a challenge of Pareto optimization, i.e., optimizing for some…

Knowledge DistillationMachine TranslationTranslation

Life Cycle-Aware Evaluation of Knowledge Distillation for Machine Translation: Environmental Impact and Translation Quality Trade-offs

2026-02-10 · Joseph Attieh, Timothee Mickus, Anne-Laure Ligozat, Aurélie Névéol 외 arxiv

Knowledge distillation (KD) is a tool to compress a larger system (teacher) into a smaller one (student). In machine translation, studies typically report only the translation quality of the student and omit the computat…

Knowledge DistillationMachine Translation

Combining Compressions for Multiplicative Size Scaling on Natural Language Tasks

2022-08-20 · COLING 2022 10 · Rajiv Movva, Jinhao Lei, Shayne Longpre, Ajay Gupta 외

Quantization, knowledge distillation, and magnitude pruning are among the most popular methods for neural network compression in NLP. Independently, these methods reduce model size and can accelerate inference, but their…

Knowledge DistillationNeural Network CompressionQuantization

ProGMLP: A Progressive Framework for GNN-to-MLP Knowledge Distillation with Efficient Trade-offs

2025-07-25 · Weigang Lu, Ziyu Guan, Wei Zhao, Yaming Yang 외 arxiv

GNN-to-MLP (G2M) methods have emerged as a promising approach to accelerate Graph Neural Networks (GNNs) by distilling their knowledge into simpler Multi-Layer Perceptrons (MLPs). These methods bridge the gap between the…

Computational EfficiencyKnowledge Distillation

Extracting General-use Transformers for Low-resource Languages via Knowledge Distillation

2025-01-22 · Jan Christian Blaise Cruz, Alham Fikri Aji

In this paper, we propose the use of simple knowledge distillation to produce smaller and more efficient single-language transformers from Massively Multilingual Transformers (MMTs) to alleviate tradeoffs associated with…

Knowledge Distillation