Machine Learning Techniques 筆記：2-8 Ada Boosting
先算出錯誤率Epselo t，把犯錯誤的點做（1-Et）的放縮，正確的點做Et比例的放縮=》正確的乘以♦t，錯誤的除以♦t 通常，♦t≥1，因爲1/2是丟銅板亂猜的概率。只要base algroithm比亂猜好，♦t大於等於1，使得正確的得以縮小，錯誤的得以放大，更加專注在犯錯誤的點上面，來得到不一樣的hypothesis。有點類似於PLA，有錯即改 只要比亂猜效果好，最終Ein總可收斂 t=1，
相關文章
- 1. Machine Learning Techniques 筆記：2-13 Deep Learning
- 2. Machine Learning Techniques 筆記：2-6 Support Vector Regression
- 3. Machine Learning Techniques 筆記：2-12 Neural Network
- 4. Machine Learning Techniques 筆記：2-15 Matrix Factorization
- 5. Machine Learning Techniques 筆記：2-5 Kernel Logistics Regression
- 6. Machine Learning Techniques 筆記：2-7 Blending and Bagging
- 7. Machine Learning Techniques 筆記：2-9 Decision Tree
- 8. Machine Learning Techniques 筆記：2-10 Random Forest
- 9. Machine Learning Techniques 筆記：2-2 Dual Support Vector Machine
- 10. Machine Learning Techniques 筆記：2-3 Kernel Support Vector Machine
- 更多相關文章...
- • Docker Machine - Docker教程
- • ASP.NET Razor - 標記 - ASP.NET 教程
- • Tomcat學習筆記（史上最全tomcat學習筆記）
- • Java Agent入門實戰（一）-Instrumentation介紹與使用
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. Machine Learning Techniques 筆記：2-13 Deep Learning
- 2. Machine Learning Techniques 筆記：2-6 Support Vector Regression
- 3. Machine Learning Techniques 筆記：2-12 Neural Network
- 4. Machine Learning Techniques 筆記：2-15 Matrix Factorization
- 5. Machine Learning Techniques 筆記：2-5 Kernel Logistics Regression
- 6. Machine Learning Techniques 筆記：2-7 Blending and Bagging
- 7. Machine Learning Techniques 筆記：2-9 Decision Tree
- 8. Machine Learning Techniques 筆記：2-10 Random Forest
- 9. Machine Learning Techniques 筆記：2-2 Dual Support Vector Machine
- 10. Machine Learning Techniques 筆記：2-3 Kernel Support Vector Machine