正確的輸出是模型對於給定的輸入應該產生的結果。

The correct output is what the model issupposed to produce for the given input.

{輸入，正確的輸出}

{ input, correct output }

有監督學習中的學習是對模型進行一系列的修改，以減少同一輸入下正確輸出與模型輸出之間的差異。

Learning in supervised learning is theseries of revisions of a model to reduce the difference between the correctoutput and the output from the model for the same input.

如果模型被完全訓練，它將產生與訓練資料的輸入相對應的正確輸出。

If a model is perfectly trained, it willproduce a correct output that corresponds to the input from the training data.

與此相反，無監督學習的訓練資料只包含沒有正確輸出的輸入。

In contrast, the training data of theunsupervised learning contains only inputs without correct outputs.

{輸入}

{ input }

乍一看，似乎很難理解，沒有正確的輸出，該如何進行訓練呢。

At a first glance, it may seem difficult tounderstand how to train without correct outputs.

然而，目前已經開發了許多這種型別的無監督學習方法。

However, many methods of this type havebeen developed already.

無監督學習通常用於研究資料的特徵和資料的預處理。

Unsupervised learning is generally used forinvestigating the characteristics of the data and preprocessing the data.

無監督學習概念類似於一個學生，他僅僅通過構造和屬性來分類問題，而沒有學習如何解決這些問題，因為沒有已知的正確輸出。

This concept is similar to a student whojust sorts out the problems by construction and attribute and doesn’t learn howto solve them because there are no known correct outputs.

強化學習採用輸入、輸出、評分等集合作為訓練資料。

Reinforcement learning employs sets ofinput, some output, and grade as training data.

強化學習通常在需要最佳互動時使用，例如控制和遊戲。

It is generally used when optimalinteraction is required, such as control and game plays.

{輸入，一些輸出，輸出評分}

{ input, some output, grade for this output}

本書只涉及監督學習。

This book only covers supervised learning.

與無監督學習和強化學習相比，監督學習的應用更為廣泛，更重要的是，它是你在進入機器學習和深度學習的世界時需要學習的第一個概念。

It is used for more applications comparedto unsupervised learning and reinforcement learning, and more importantly, itis the first concept you will study when entering the world of Machine Learningand Deep Learning.

分類與迴歸（Classification and Regression）

有監督學習的兩種最常用型別是分類和迴歸。

The two most common types of application ofsupervised learning are classification and regression.

這些術語聽起來可能不熟悉，但實際上並不那麼具有挑戰性。

These words may sound unfamiliar, but areactually not so challenging.

我們從分類開始談起。

Let’s start with classification.

這可能是機器學習中最普遍的應用。

This may be the most prevailing applicationof Machine Learning.

從字面上來說，分類問題的重點是找到資料所屬的類別。

The classification problem focuses onliterally finding the classes to which the data belongs.

舉例說明可能會有所幫助。

Some examples may help.

垃圾郵件過濾服務➔將郵件分類為普通郵件或垃圾郵件

Spam mail filtering service ➔ Classifies the mails by regular or spam

數字識別服務➔將數字影象分類為0—9

Digit recognition service ➔ Classifies the digit image into one of 0-9

人臉識別服務➔將人臉影象分類為註冊使用者之一

Face recognition service ➔ Classifies the face image into one of the registered users

我們在上一節中談到，有監督學習需要用於訓練資料的由輸入和正確輸出組成的輸入輸出對。

We addressed in the previous section thatsupervised learning requires input and correct output pairs for the trainingdata.

——本文譯自Phil Kim所著的《Matlab Deep Learning》

更多精彩文章請關注微訊號：