Glossary math Term Page

Least Squares

A best-fit method based on squared residuals

least-squares #math#linear-algebra

Korean version

Aliases

least squares

Prerequisites

Projection Dot Product

Related Concepts

Projection Column Space Linear Regression

Core Idea

Least squares chooses an approximate solution by minimizing the squared Euclidean norm of the residual. It is the standard response when an exact solution to Ax=b does not exist.

Role In This Blog

In Mathbong, least squares connects projection, residuals, normal equations, and linear regression.

Posts Mentioning This Concept

[Linear Algebra Series Part 5] Dot Product and Cosine Similarity Understand the dot product as both component contribution and directional alignment, then connect it to cosine similarity.[Linear Algebra Series Part 18] Least Squares and the Basics of Linear Regression Use [[least-squares|least squares]] and [[linear-regression|linear regression]] to find the best approximate solution when `Ax = b` has no exact answer.[선형대수 시리즈 18편] 최소제곱과 선형회귀의 기초 해가 정확히 맞지 않는 선형시스템에서 최소제곱 해를 구하는 아이디어를 선형회귀와 연결합니다.