1: \begin{abstract}

2: The randomized Kaczmarz (RK) method is a well-known approach for solving linear least-squares problems with a large number of rows.

3: RK accesses and processes just one row at a time, leading to exponentially fast convergence for consistent linear systems.

4: However, RK fails to converge to the least-squares solution for inconsistent systems.

5: This work presents a simple fix: average the RK iterates produced in the tail part of the algorithm.

6: The proposed tail-averaged randomized Kaczmarz (TARK) converges for both consistent and inconsistent least-squares problems at a polynomial rate, which is known to be optimal for any row-access method.

7: An extension of TARK also leads to efficient solutions for ridge-regularized least-squares problems.

8: \end{abstract}

9: