paper-with-me

홈 › Papers

An Evaluation of Zero-Cost Proxies -- from Neural Architecture Performance to Model Robustness

2023-07-18 · Jovita Lukasik, Michael Moeller, Margret Keuper

Zero-cost proxies are nowadays frequently studied and used to search for neural architectures. They show an impressive ability to predict the performance of architectures by making use of their untrained weights. These techniques allow for immense search speed-ups. So far the joint search for well-performing and robust architectures has received much less attention in the field of NAS. Therefore, the main focus of zero-cost proxies is the clean accuracy of architectures, whereas the model robustness should play an evenly important part. In this paper, we analyze the ability of common zero-cost proxies to serve as performance predictors for robustness in the popular NAS-Bench-201 search space. We are interested in the single prediction task for robustness and the joint multi-objective of clean and robust accuracy. We further analyze the feature importance of the proxies and show that predicting the robustness makes the prediction task from existing zero-cost proxies more challenging. As a result, the joint consideration of several proxies becomes necessary to predict a model's robustness while the clean accuracy can be regressed from a single such feature.

📄 PDF Abstract BibTeX arXiv:2307.09365

Code (0)

등록된 구현이 없습니다.

Tasks

Feature Importance

Methods 이 논문이 사용한 방법론

Focus 설명 없음

Similar Papers 제목 키워드 기반

Zero-Shot NAS via the Suppression of Local Entropy Decrease

2024-11-09 · Ning Wu, Han Huang, Yueting Xu, Zhifeng Hao

Architecture performance evaluation is the most time-consuming part of neural architecture search (NAS). Zero-Shot NAS accelerates the evaluation by utilizing zero-cost proxies instead of training. Though effective, exis…

Neural Architecture Search

ParZC: Parametric Zero-Cost Proxies for Efficient NAS

2024-02-03 · Peijie Dong, Lujun Li, Xinglin Pan, Zimian Wei 외

Recent advancements in Zero-shot Neural Architecture Search (NAS) highlight the efficacy of zero-cost proxies in various NAS benchmarks. Several studies propose the automated design of zero-cost proxies to achieve SOTA p…

Neural Architecture Search

A Deeper Look at Zero-Cost Proxies for Lightweight NAS

2022-01-17 · ICLR Track Blog 2022 5 · Anonymous

While early algorithms for neural architecture search (NAS) used extreme computational budgets, recent techniques have aimed to lower the computation time. Very recently, a new family of techniques were introduced to app…

Neural Architecture Search

GreenMachine: Automatic Design of Zero-Cost Proxies for Energy-Efficient NAS

2024-11-22 · Gabriel Cortês, Nuno Lourenço, Penousal Machado

Artificial Intelligence (AI) has driven innovations and created new opportunities across various sectors. However, leveraging domain-specific knowledge often requires automated tools to design and configure models effect…

Neural Architecture Search

EZNAS: Evolving Zero Cost Proxies For Neural Architecture Scoring

2022-09-15 · Yash Akhauri, J. Pablo Munoz, Nilesh Jain, Ravi Iyer

Neural Architecture Search (NAS) has significantly improved productivity in the design and deployment of neural networks (NN). As NAS typically evaluates multiple models by training them partially or completely, the impr…

Neural Architecture Search