Lasso Regression is similar to ridge regression, but instead we use the L1 regularization penalty. This has the effect of setting some coefficient estimates to exactly zero for the least influential variables, which leads to models only including a subset of variables. This is called a space solution, which is a kind of feature selection. The parameter $\alpha$ controls the amount of L1 regularization (default = 1.0). The prediction formula is the same as OLS.