Udacity Deep Learning Course By Google


Deep Learning By Google @ udacity

FYI, the course link is https://www.udacity.com/course/deep-learning–ud730. This course takes Approximately 3 months with assumption 6hrs/wk (work at your own pace).

Knowledge Needed

  • Python (using jupyter notebook)

Assumption On Achievement

  • Good at using Tensorflow: which should be the main reason google had built this course, to promote their open source machine learning framework. IMHO, Tensorflow really lack of documentation/example due to its age.
  • More deep learning knowledge: Andrew Ng Machine Learning course only slightly covered neural network, I do hope can learn something new about deep learning here


This course consist of four Lessons:

  • Lesson 1: From Machine Learning to Deep Learning
  • Lesson 2: Deep Neural Networks
  • Lesson 3: Convolutional Neural Networks
  • Lesson 4: Deep Models for Text and Sequences

So far on Lessons 1

Nothings new if you had attended other machine learning course. Basicly:

  • Prepare your Docker environment which consist of Course Materials with Jupyter Notebook for the future Lesson
  • Walkthrough on basic machine learning rules such as learning rate, overfitting, etc.


This is really a course that machine learning guys shouldn’t miss. Deep learning is the state of the art machine learning technique which has been used to solve many real world problem.

You probably know that AlphaGo - which is mainly written in deep learning - has beaten the three-time European Go champion Fan Hui 5–0.

Written on February 7, 2016