Study towards the proposal of a new Zero-shot NN evaluation index
-
- TAKAHASHI Chisato
- Tokyo City University
-
- KENYA Jin'no
- Tokyo City University
Bibliographic Information
- Other Title
-
- 新たなZero-shot NN評価指標の提案に向けた検討
Abstract
<p>Neural Architecture Search (NAS), which automatically optimizes the structure of neural networks, garner attention in recent years. NAS has the problem that it takes an enormous amount of time to search. For this reason, a zero-shot evaluation method has been proposed to estimate classification accuracy without training. The previously proposed zero-shot indices assess performance by defining expression based on the activity of the output or the derivative of the output with respect to the input. However, these indices tend to overestimate the performance of Neural Networks with a high number of parameters. We therefore decid to investigate whether there is a way to solve this problem. We observe that the robustness of the ReLU output distribution with respect to the weights increases when the performance of the neural network is high.</p>
Journal
-
- Proceedings of the Annual Conference of JSAI
-
Proceedings of the Annual Conference of JSAI JSAI2023 (0), 4Xin179-4Xin179, 2023
The Japanese Society for Artificial Intelligence
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1390296808221582080
-
- ISSN
- 27587347
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
-
- Abstract License Flag
- Disallowed