神經網絡不同激活函數比較--讀《Understanding the difficulty of training deep feedforward neural networks》
轉載：https://ziyubiti.github.io/2016/11/01/actfunpaper/ 這篇論文比較了不同激活函數的中間各隱層輸出值分佈、梯度值分佈、測試準確度等，從上圖看出，至少在mnist和cifar10測試集下，看起來tanh和softsign都要好於sigmoid。
相關文章
- 1. Xavier——Understanding the difficulty of training deep feedforward neural networks
- 2. 【Deep Learning】筆記：Understanding the difficulty of training deep feedforward neural networks
- 3. 論文解析-《Understanding the difficulty of training deep feedforward neural networks》
- 4. Paper之DL之BP：《Understanding the difficulty of training deep feedforward neural networks》
- 5. 神經網絡剖析激活函數優缺點（Undetanding the difficulty of training deep feedforward neural networks）
- 6. [論文筆記] [2010] Understanding the Difficulty of Training Deep Feedforward Neural Networks
- 7. On the difficulty of training Recurrent Neural Networks
- 8. 理解訓練深層前饋神經網絡的難度（Undetanding the difficulty of training deep feedforward neural networks ）...
- 9. 人工神經網絡之激活函數 -Sigmoid函數
- 10. 人工神經網絡之激活函數 -tanh函數
- 更多相關文章...
- • PHP 類型比較 - PHP教程
- • 網絡協議是什麼？ - TCP/IP教程
- • JDK13 GA發佈：5大特性解讀
- • Flink 數據傳輸及反壓詳解
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. Xavier——Understanding the difficulty of training deep feedforward neural networks
- 2. 【Deep Learning】筆記：Understanding the difficulty of training deep feedforward neural networks
- 3. 論文解析-《Understanding the difficulty of training deep feedforward neural networks》
- 4. Paper之DL之BP：《Understanding the difficulty of training deep feedforward neural networks》
- 5. 神經網絡剖析激活函數優缺點（Undetanding the difficulty of training deep feedforward neural networks）
- 6. [論文筆記] [2010] Understanding the Difficulty of Training Deep Feedforward Neural Networks
- 7. On the difficulty of training Recurrent Neural Networks
- 8. 理解訓練深層前饋神經網絡的難度（Undetanding the difficulty of training deep feedforward neural networks ）...
- 9. 人工神經網絡之激活函數 -Sigmoid函數
- 10. 人工神經網絡之激活函數 -tanh函數