paper-with-me

홈 › Papers

NAS-LID: Efficient Neural Architecture Search with Local Intrinsic Dimension

2022-11-23 · Xin He, Jiangchao Yao, Yuxin Wang, Zhenheng Tang, Ka Chu Cheung, Simon See, Bo Han, Xiaowen Chu

One-shot neural architecture search (NAS) substantially improves the search efficiency by training one supernet to estimate the performance of every possible child architecture (i.e., subnet). However, the inconsistency of characteristics among subnets incurs serious interference in the optimization, resulting in poor performance ranking correlation of subnets. Subsequent explorations decompose supernet weights via a particular criterion, e.g., gradient matching, to reduce the interference; yet they suffer from huge computational cost and low space separability. In this work, we propose a lightweight and effective local intrinsic dimension (LID)-based method NAS-LID. NAS-LID evaluates the geometrical properties of architectures by calculating the low-cost LID features layer-by-layer, and the similarity characterized by LID enjoys better separability compared with gradients, which thus effectively reduces the interference among subnets. Extensive experiments on NASBench-201 indicate that NAS-LID achieves superior performance with better efficiency. Specifically, compared to the gradient-driven method, NAS-LID can save up to 86% of GPU memory overhead when searching on NASBench-201. We also demonstrate the effectiveness of NAS-LID on ProxylessNAS and OFA spaces. Source code: https://github.com/marsggbo/NAS-LID.

📄 PDF Abstract BibTeX arXiv:2211.12759

Code (1)

marsggbo/nas-lid 공식 구현 pytorch

Tasks

GPUNeural Architecture Search

Methods 이 논문이 사용한 방법론

Adam 설명 없음
DropPath Just as dropout prevents co-adaptation of activations, DropPath prevents co-adaptation of parallel paths in networks such as…
REINFORCE REINFORCE is a Monte Carlo variant of a policy gradient algorithm in reinforcement learning. The agent collects samples of an episode using its current policy, and uses it to…
Cutout Cutout is an image augmentation and regularization technique that randomly masks out square regions of input during training. and can be used to improve the robustness and…
ProxylessNAS ProxylessNAS directly learns neural network architectures on the target task and target hardware without any proxy task. Additional contributions include: - Using a new…
OFA In this work, we pursue a unified paradigm for multimodal pretraining to break the scaffolds of complex task/modality-specific customization. We propose OFA, a Task-Agnostic and…

Similar Papers 제목 키워드 기반

ABID: Angle Based Intrinsic Dimensionality

2020-06-23 · Erik Thordsen, Erich Schubert

The intrinsic dimensionality refers to the ``true'' dimensionality of the data, as opposed to the dimensionality of the data representation. For example, when attributes are highly correlated, the intrinsic dimensionalit…

Dimensionality Reduction

Local Intrinsic Dimension of Representations Predicts Alignment and Generalization in AI Models and Human Brain

2026-01-30 · Junjie Yu, Wenxiao Ma, Chen Wei, Jianyu Zhang 외 arxiv

Recent work has found that neural networks with stronger generalization tend to exhibit higher representational alignment with one another across architectures and training paradigms. In this work, we show that models wi…

Intrinsic Dimensionality Estimation within Tight Localities: A Theoretical and Experimental Analysis

2022-09-29 · Laurent Amsaleg, Oussama Chelly, Michael E. Houle, Ken-ichi Kawarabayashi 외

Accurate estimation of Intrinsic Dimensionality (ID) is of crucial importance in many data mining and machine learning tasks, including dimensionality reduction, outlier detection, similarity search and subspace clusteri…

Dimensionality ReductionOutlier Detection

Shared Global and Local Geometry of Language Model Embeddings

2025-03-27 · Andrew Lee, Melanie Weber, Fernanda Viégas, Martin Wattenberg

Researchers have recently suggested that models share common representations. In this work, we find that the token embeddings of language models exhibit common geometric structure. First, we find ``global'' similarities:…

Language ModelingLanguage Modelling

k-NN Regression Adapts to Local Intrinsic Dimension

2011-12-01 · NeurIPS 2011 12 · Samory Kpotufe

Many nonparametric regressors were recently shown to converge at rates that depend only on the intrinsic dimension of data. These regressors thus escape the curse of dimension when high-dimensional data has low intrinsi…

regression