Machine Learning——Lecture 6
爲什麼邏輯迴歸不能用平方誤差做損失函數，而要用交叉熵函數？ 爲什麼生成式要比判別式的準確率低？ 以爲生成式要自學，它在樣本少的時候將明顯是class1的樣本誤分到class2中。 多分類任務softmax如何計算分類結果？ 邏輯迴歸的侷限（由於boundary只是一條直線） 做了transformation之後就可以區分開了，但是難點在於往哪個方向做transformation
相關文章
- 1. Lecture 2: Supervised machine learning
- 2. Machine learning(6): Quiz 6
- 3. Machine learning(6): SVMs
- 4. Coursera Machine Learning Week 6 - Advice for Applying Machine Learning
- 5. ML Lecture 0：Introduction of Machine Learning / Why we need to learn machine learning?
- 6. Andrew Ng 's machine learning lecture note (8)
- 7. Andrew Ng 's machine learning lecture note (11)
- 8. Andrew Ng 's machine learning lecture note (14)
- 9. Andrew Ng 's machine learning lecture note (3)
- 10. ML Lecture 0-1: Introduction of Machine Learning
- 更多相關文章...
- • Docker Machine - Docker教程
- • Redis的6種數據類型 - Redis教程
- • Java Agent入門實戰（一）-Instrumentation介紹與使用
- • Java Agent入門實戰（三）-JVM Attach原理與使用
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. Lecture 2: Supervised machine learning
- 2. Machine learning(6): Quiz 6
- 3. Machine learning(6): SVMs
- 4. Coursera Machine Learning Week 6 - Advice for Applying Machine Learning
- 5. ML Lecture 0：Introduction of Machine Learning / Why we need to learn machine learning?
- 6. Andrew Ng 's machine learning lecture note (8)
- 7. Andrew Ng 's machine learning lecture note (11)
- 8. Andrew Ng 's machine learning lecture note (14)
- 9. Andrew Ng 's machine learning lecture note (3)
- 10. ML Lecture 0-1: Introduction of Machine Learning