NTD in AI: Bias

Non-technical definitions in AI

Bias is the difference between the relationship between the data we are studying in our statistical model and the real world.

In the real world, the relationship between a set of variables are usually hidden and more complex than the models we train for statistical inference. For example say we apply a simple linear regression model to a problem but in the real world, the relationship is actually more complex and non-linear, the model would have a high bias.

Intuitively we can imagine that the less “flexible” a model is (such as linear regression, since we are trying to fit a straight line or plane/spline to a scatter of data points), the higher the bias.

Machine learning is a technical subject and the use of technical terms by engineers have the potential of coming between clear communication with non-engineers, especially in the business setting. In spare moments I started to put together simple, non-technical definitions of nouns and verbs used in the field of machine learning as a kind of Rosetta Stone for non-engineers.This is a work-in-progress which I may collect into a book one day. This is one of those definitions.

NTD in AI: Bias

Other non-technical definitions: