Coursera’s “Machine Learning for Everyone” Fulfills Unmet Training Requirements « Machine Learning Times – The Predictive Analytics Times

My new course series on Coursera, Machine Learning for Everyone (free access), fulfills two different kinds of unmet learner needs. It’s a conceptually-complete, end-to-end course series – its three courses amount to the equivalent of a college or graduate-level course – that covers both the technology side and the business side. While fully accessible and understandable to business-level learners, it’s also also vital to data scientists and budding technical practitioners, since it covers:

  • The state-of-the-art techniques
  • The business leadership best practices
  • A wide range of common pitfalls and how to avoid them

1) A comprehensive go-to for BUSINESS-SIDE learners – by covering the following:

  • ML project leadership (management process)
  • ML algorithms: substantive yet accessible coverage
  • Data preparation

2) Need-to-knows for EVERYONE in ML – both business-side learners and technical practitioners – by also covering the following:

  • ML ethics: risks to social justice, equitable models, machine bias, etc.
  • Business-oriented performance metrics
  • Uplift modeling (aka persuasion modeling)
  • Major pitfalls, in-depth:
    • P-hacking
    • Overfitting
    • The accuracy fallacy
    • Presuming causation from correlations
    • Serious problems with hyping ML as “AI”

This checklist illustrates the unique contribution of this curriculum:

More information about “Machine Learning for Everyone”:

Brief curriculum overview (video)

Seven Reasons Budding Data Scientists Need