2
2

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

More than 1 year has passed since last update.

ICML2022の枝刈り論文 (1)

Last updated at Posted at 2022-07-04

概要

この記事では、ICML2022の枝刈り論文を紹介します。1

PAC-Net: A Model Pruning Approach to Inductive Transfer Learning

概要:PAC-Net. Inductive Transfer Learning (10~100など非常に少量データでの転移学習)の手法。
研究機関:Samsung
新規性:Inductive Transfer LearningでSOTA。
キモ:Pretrained modelを刈って空いた所に新しいタスクを学習させる。その際正則化項をつける。
image.png
image.png
微分方程式でバネ・マス・ダンパ系を学習する例だと、基本的な運動をsourceで、ダンパの寄与をtargetで学習する。
評価:微分方程式や回帰問題で実験した。

  1. 画像や数式は論文から引用しています。

2
2
0

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
2
2

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?