DAY 3
1

# Course - How Google does Machine Learning

• What it means to be AI first
• `What it means to be AI first`
• `Two stages of ML`
• `ML in Google products`
• `Demo: ML in Google products`
• `Demo: ML in Google Photos`
• `Google Translate and Gmail`
• `Replacing heuristics`
• Framing an ML problem
• Lab Intro - Framing an ML Problem
• Framing an ML problem
• Lab debrief
• ML in applications
• Demo: ML in applications
• Pre-trained models
• The ML marketplace is evolving
• A data strategy
• Training and serving skew
• An ML strategy
• Lab Intro - ML use case
• Module 2 Quiz

## 1. What it means to be AI first

• What it means to be AI first
• What it means to be AI first

AI是一種discipline(學科)，有許多的theory and methods

ML是一種toolset(工具集)，ML就是拿來解AI問題的
(ML是讓machine訓練與學習用的，他們沒有intelligent，但會become intelligent)

## 2. Two stages of ML

• What it means to be AI first
• Two stages of ML

### ML的兩個stages:

• Stage1 - training
• Stage2 - inference(推理)(也有些人稱為prediction(預測))

### Stage1 : training

• input： input資料本身
• label： input資料的所對應的答案 (見下圖左邊)

ML model的本身就是一種數學函數(math functions)

ML 需要的就是 labeled examples = input + label(ture anwser)

## 3. ML in Google products

• What it means to be AI first

• 原因1在於電腦運算能力

• 原因2在於資料的數量

• 原因3在於計算技巧

One solution needs many ML models.

## 4. Google Translate and Gmail

• What it means to be AI first
• Demo: ML in Google products

## 5. Replacing heuristics

• What it means to be AI first
• Replacing heuristics

"Machine learning," says Eric, "This is the next transformation. The programming paradigm is changing. It's not programming a computer. You teach a computer to learn something and then it does what you want."

ML是一種logic，而並非只是在處理data。

google要怎麼知道我們想搜尋的是 San Francisco Giants 還是 New York Giants 呢?

ML還能夠替我們解決什麼問題? 如果提到 ML 未來的可能性，

Notice that saying that machine learning is a way to replace rules, notice that this is a far more expansive answer to what kinds of problems can machine learning solve.

(這段真的太猛了...，如果真的實現了等於不用再慢慢地針對情況一一對應coding了，