---
title: Machine Learning
tags: APAC HPC-AI competition
---

# Machine Learning
[TOC]

---

誰來幫影片們上個華麗麗的語法
nobody
真是狠心

[如果你很閒可以送他3分鐘(開2倍速)](https://www.youtube.com/watch?v=HcqpanDadyQ)

用舉例的方式說明machine learning，不是我們習慣的腔調但有字幕我相信大家可以(如果真的不行請求助miss Chen)
[Machine Learning Basics](https://www.youtube.com/watch?v=ukzFI9rgwfU&t=17s)

deeplearning速成班 不用字幕也聽得懂
[Deep Learning In 5 Minutes](https://www.youtube.com/watch?v=6M5VXKLf4D4)

[我怎麼就來了資工系勒第一集](https://www.youtube.com/watch?v=aircAruvnKk)

[第二集 大腦大腦真奇妙](https://www.youtube.com/watch?v=IHZwWFHWa-w)

有中文字幕就推!!

---
## What is Machine Learning?
Wikipedia’s definition:

>Machine learning is a scientific discipline that explores the construction and study of algorithms that can learn from data.

Although an advanced topic, Machine Learning is growing to be one of the most in-demand sciences. With the immense amount of data in the world today, every aspect of life has potential to be improved through Machine Learning.

Below are some applications of Machine Learning:

1. Stock Trading and Market Analysis
2. Recommender Systems (Netflix, Pandora, Spotify, etc.)
3. Natural Language Processing (IBM Watson)
4. Bioinformatics (Study of genes that increase likelihood of Cancer)
5. Computer Vision (Self-driving Car)
### How do computers discover new knowledge?
1. Fill in gaps in existing knowledge
2. Emulate the brain
3. Simulate evolution
4. Systematically reduce uncertainty
5. Notice similarities between old and new

## [The Five Tribes of Machine Learning](https://www.youtube.com/watch?time_continue=608&v=E8rOVwKQ5-8&feature=emb_logo)

(Tribe/Origins/Master Algorithm):
> tribe 也可翻譯為 (分類中一個階級的)族、類別
1. **Symbolists/Logic, philosophy/Inverse deduction**: They don't start with a premise to work towards conclusions, but rather use a set of premises and conclusions and work backward to fill in the gaps.
2. **Connectionists/Neuroscience/Backpropagation**: They mostly try to digitally re-engineer the brain and all of its connections in a neural network. The most famous example of the connectionist approach is what is commonly known as ‘Deep Learning’. Their techniques have proved very efficient in e.g. image recognition and machine translation.
3. **Evolutionaries/Evolutionary biology/Genetic programming**: Their focus lies on applying the idea of genomes and DNA in the evolutionary process to data processing: their algorithms will constantly evolve and adapt to unknown conditions and processes.
4. **Bayesians/Statistics/Probabilistic inference**: Bayesian models will take a hypothesis and apply a type of "a priori" thinking, believing that there will be some outcomes that are more likely. They then update their hypothesis as they see more data.
5. **Analogizers/Psychology/Kernel machines**: This machine learning discipline focuses on techniques to match bits of data to each other. Probably the most famous example of this type of machine learning, is the Amazon or Netflix recommendations: "If you have watched/bought this, you will; probably like..."

### [TensorFlow: A Framework for Scalable Machine Learning](https://www.youtube.com/watch?time_continue=576&v=7meUnPoFBIs&feature=emb_logo)
>Tensor --> A multidimensional array
>Flow --> A graph of operations

#### What does TensorFlow handle?
1. Modeling complexity
2. Distributed system
3. Heterogeneous system