arXiv cs.AI· Nikita Gabdullin, Anton Raskovalov·· 13 小时前AI 评分22
利用潜在空间配置加速神经网络嵌入的Top-K检索
Latent space configuration for neural network finetuning, fast ranking, and embedding dimensionality reduction
AI 导读
研究者提出面向神经网络结构化潜在空间的搜索算法，通过O(1)复杂度的小候选集搜索加速Top-K检索。基于ArcFace预训练的IResNet50模型用Latent Space Configuration（LSC）方法微调后，相比基线穷举搜索在GPU上最高提速3倍、CPU上提速40%。潜在空间降维还能进一步加速搜索并提升rank-N指标。
正文 · 原文
Abstract:Finding K most similar elements for a query in a large gallery of neural network (NN) embeddings is essential in various areas of computer science and industrial applications of artificial intelligence. This paper proposes search algorithms for structured latent spaces (LS) of NNs which allow to speed up top K item retrieval owing to O(1) complexity small candidate gallery subset search. ArcFace-pretrained IResNet50 models are finetuned using Latent Space Configuration (LSC) methodology to obtain new distribution of embeddings with desired geometric properties. Applying the proposed algorithms to embeddings obtained from finetuned models leads to several times speed-up over baseline exhaustive search. Specifically, we report up to 3 times speed-up on GPU and 40% speed-up on CPU compared to the baseline method GPU speed. These results are obtained for the original and reduced numbers of LS dimensions. The results show that LS dimensionality reduction allows to further accelerate the search and even improve rank-N metrics.
| Comments: | 14 pages, 4 figures, 2 tables, 2 algorithms |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.29295 [cs.CV] |
| (or arXiv:2610.29295v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.29295 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Nikita Gabdullin [view email]
[v1]
Wed, 21 Oct 2026 06:47:13 UTC (321 KB)
来源：arXiv cs.AI · arxiv.org