Sammenlign metoder
Gjennomgå de valgte metodene side om side; rader som avviker, er uthevet.
| Online lineær regresjon× | Regularisert lineær regresjon× | |
|---|---|---|
| Fagfelt | Maskinlæring | Maskinlæring |
| Familie | Machine learning | Machine learning |
| Opprinnelsesår≠ | 1960 (LMS); 1950 (RLS formalization) | 1970–2005 |
| Opphavsperson≠ | Widrow, B. & Hoff, M. E. (LMS); Gauss / Plackett (RLS) | Hoerl & Kennard (Ridge, 1970); Tibshirani (Lasso, 1996); Zou & Hastie (Elastic Net, 2005) |
| Type≠ | Incremental supervised regression | Penalized linear model |
| Opprinnelig kilde≠ | Shalev-Shwartz, S. (2012). Online Learning and Online Convex Optimization. Foundations and Trends in Machine Learning, 4(2), 107–194. DOI ↗ | Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B, 58(1), 267–288. DOI ↗ |
| Alias | incremental linear regression, streaming linear regression, recursive least squares regression, stochastic gradient descent regression | Ridge regression, Lasso regression, Elastic Net regression, penalized regression |
| Relaterte≠ | 6 | 4 |
| Sammendrag≠ | Online Linear Regression fits a linear model one observation at a time, updating weights incrementally as each new data point arrives. Unlike batch least-squares, it never needs to store or re-process the full dataset, making it the natural choice for streaming data, very large datasets, and environments where the data-generating process can shift over time. | Regularized linear regression adds a penalty term to the ordinary least-squares objective, shrinking or zeroing out coefficients to reduce overfitting and handle multicollinearity. The three main variants — Ridge (L2 penalty), Lasso (L1 penalty), and Elastic Net (combined L1+L2) — make linear regression usable even when features outnumber observations or predictors are highly correlated. |
| ScholarGateDatasett ↗ |
|
|