Most important concepts in applied machine learning.

Image for post
Image for post
Ideas in this post aren’t as shiny as this lightbulb, but they do their job better than alternatives.

1. Scale drives machine learning progress

Image for post
Image for post

2. Optimization of algorithms

Image for post
Image for post
Two algorithms with different properties

2.1 Optimize and satisfy

Evaluation metric = Accuracy * (Delay < 0.5s)
Image for post
Image for post

2.2 Single-number evaluation metric

Evaluation metric = Accuracy - (Delay / 100)
Image for post
Image for post
With this formulation, Algorithm C doesn’t get discarded

2.3 Person makes the call

3. Development

“Do you already have the data, or can we obtain it easily?”

3.1 Specific examples and metrics-level data

Image for post
Image for post
Macro data. Gives weight to different problems.
Image for post
Image for post
Full development cycle. Showing micro and macro data are related, author: Taivo Pungas

3.2 Data is the specification

3.3 Statistical unit testing

Unit tests are typically automated tests written and run by software developers to ensure that a section of an application (known as the “unit”) meets its design and behaves as intended.

3.4 Statistical integration testing

Organiser of people, builder of software.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store