AutoML in Practice - Online


Machine learning has achieved considerable successes in recent years. However, this success greatly relies on human experts to perform many tasks, like Data cleaning, Features construaction, Optimize model hyperparameters, Analyze results, etc.

This is why Automated Machine Learning(AutoML and AutoDL) are increasingly valued by academics and industry. The core of AI is the model design. AutoML reduces the barriers to AI application, enabling developers with no AI expertise to independently and easily develop and deploy AI models.

This crash course provides you a clear picture of technical details behind the state-of-the-art AutoML, from feature engineering, feature reduction to deep learning model architecture design. This course will also discuss time series modeling. In the end, we will work together on a hands-on project using AutoML, therefore you not only learn the knowledge but apply it to solve real-world applications as well

  • Session 1: Jan. 29th Tue 10am-12pm PT
  • Session 2: Jan. 31st Thu 10am-12pm PT
  • Session 3: Feb. 5th Tue 10am-12pm PT
  • Session 4: Feb. 7th Thu 10am-12pm PT

  • 32 topics / 8 hours live talks
  • 1 code lab/ 1 hands-on competition
  • Real time interactions with instructors
  • Watch recorded videos any time

Check the content tab for full course outlines.

Developers with interests in building large scale machine learning systems. Both beginners and experts of the field can learn from different perspectives

basic knowledge of machine learning

Not sure if the course right for you? Try the first session for free. we will refund you before Jan.30 when the 2nd session starts.

If you miss the live session or want to learn again, you can watch recorded lessons any time.
Also live discuss with instructors and other developers at slack group.

Module 1: Introduction to AutoML
  • Case study: What is AutoML.
  • The components of Machine Learning and how AutoML functions
  • AutoML in traditional Machine Learning
  • AutoML in deep learning

Module 2: Deep dive into AutoML
  • Demystify state-of-art AutoDL technologies
  • Model search algorithm
  • Morphism algorithm
  • Differentiable architecture search
  • Reinforcement learning based architecture search

Module 3: Time Series Forecasting
  • Introduction to time series forecasting
  • Statistics based forecasting models (AR, MA, SARIMA)
  • Traditional machine learning based forecasting models
  • Build forecasting models using above two methods on a historical product sales data

Module 4: Deep learning time series forecasting
  • RNN, LSTM, CNN, and WaveNet
  • Limitations and challenges
  • Use AutoML

Module 5: Hands on competition on time series forecasting
  • code lab/project
  • next steps
Yuan Shen, Ning Jiang

Yuan Shen is the founder and CEO of, the first Automatic Deep Learning platform that automatically builds, trains, and deploys AI models. Yuan has more than a decade of experience in Machine Learning across medical imaging, e-commerce, online advertising, and search engines. Prior to, Yuan worked at Microsoft and eBay.

Ning Jiang CTO of Ning has over 15 years of experience in Machine Learning across multiple industries, including web/entity search, search ads, online retail, and cybersecurity. His was a senior manager at Microsoft responsible for Bing Maps Autosuggest, and managed Bing Local Search in France and Italy and the Restaurant segment. Before this, he led the development of Bing Ads relevance platform and algorithm improvements, delivering paid search experience in both Bing and Yahoo!. Ning also had experience in developing malware detection and spam detection algorithms during his tenure with China top security firm.

  • Start Date: On-demand
  • Venue: Online
  • Fee:
    $79 $9
  • Students enrolled:197
  • Status: learn on-demand
  • Preview this course:
Enroll This Course