Bias and Variance
o1 Pro.icon bias (Bias) is the degree to which something is systematically deviated from its true or ideal value when it is measured or estimated. Variance (Variance) is the degree to which the results of the same measurement or estimation can vary when repeated many times. Bias: How far off the average
Variance: the amount of variation in measurement or estimation results
In the darts example, the bias is "which direction the arrow is biased" and the variance is "how much it is scattered". This is used not only in statistical analysis and machine learning, but also as a general concept when evaluating something.
https://gyazo.com/e31295b3bd913e88c25696ee65ecbcd7
A picture of a dart you see often.
https://gyazo.com/2830d5ef710e5a3324225b36ffc1fb15
This is going to be confusing because the same term pair is used in different ways.
https://gyazo.com/109ac2fd71221157ae91ef6e04db6290
I didn't know this.
---
This page is auto-translated from /nishio/バイアスとバリアンス using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.