paper-with-me

홈 › Papers

Condensed Sample-Guided Model Inversion for Knowledge Distillation

2024-08-25 · Kuluhan Binici, Shivam Aggarwal, Cihan Acar, Nam Trung Pham, Karianto Leman, Gim Hee Lee, Tulika Mitra

Knowledge distillation (KD) is a key element in neural network compression that allows knowledge transfer from a pre-trained teacher model to a more compact student model. KD relies on access to the training dataset, which may not always be fully available due to privacy concerns or logistical issues related to the size of the data. To address this, "data-free" KD methods use synthetic data, generated through model inversion, to mimic the target data distribution. However, conventional model inversion methods are not designed to utilize supplementary information from the target dataset, and thus, cannot leverage it to improve performance, even when it is available. In this paper, we consider condensed samples, as a form of supplementary information, and introduce a method for using them to better approximate the target data distribution, thereby enhancing the KD performance. Our approach is versatile, evidenced by improvements of up to 11.4% in KD accuracy across various datasets and model inversion-based methods. Importantly, it remains effective even when using as few as one condensed sample per class, and can also enhance performance in few-shot scenarios where only limited real data samples are available.

📄 PDF Abstract BibTeX arXiv:2408.13850

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge DistillationmodelNeural Network CompressionTransfer Learning

Similar Papers 제목 키워드 기반

CD^2: Constrained Dataset Distillation for Few-Shot Class-Incremental Learning

2026-01-13 · Kexin Bao, Daichi Zhang, Hansong Zhang, Yong Li 외 arxiv

Few-shot class-incremental learning (FSCIL) receives significant attention from the public to perform classification continuously with a few training samples, which suffers from the key catastrophic forgetting problem. E…

Few-Shot Class-Incremental Learning

Small Scale Data-Free Knowledge Distillation

2024-06-12 · CVPR 2024 1 · He Liu, Yikai Wang, Huaping Liu, Fuchun Sun 외

Data-free knowledge distillation is able to utilize the knowledge learned by a large teacher network to augment the training of a smaller student network without accessing the original training data, avoiding privacy, se…

Data-free Knowledge DistillationGenerative Adversarial Networkimage-classificationImage Classification+2

Feature-Rich Audio Model Inversion for Data-Free Knowledge Distillation Towards General Sound Classification

2023-03-14 · Zuheng Kang, Yayun He, Jianzong Wang, Junqing Peng 외

Data-Free Knowledge Distillation (DFKD) has recently attracted growing attention in the academic community, especially with major breakthroughs in computer vision. Despite promising results, the technique has not been we…

Data-free Knowledge DistillationKnowledge DistillationSound Classification

Safeguard Text-to-Image Diffusion Models with Human Feedback Inversion

2024-07-17

This paper addresses the societal concerns arising from large-scale text-to-image diffusion models for generating potentially harmful or copyrighted content. Existing models rely heavily on internet-crawled data, wherein…

IMS3: Breaking Distributional Aggregation in Diffusion-Based Dataset Distillation

2026-03-14 · Chenru Wang, Yunyi Chen, Zijun Yang, Joey Tianyi Zhou 외 arxiv

Dataset Distillation aims to synthesize compact datasets that can approximate the training efficacy of large-scale real datasets, offering an efficient solution to the increasing computational demands of modern deep lear…