1: \begin{abstract}

2: We consider the problem of reconstructing rank-one matrices from random linear measurements, a task that appears in a variety of problems in signal processing, statistics, and machine learning.

3: In this paper, we focus on the Alternating Least Squares (ALS) method. 

4: While this algorithm has been studied in a number of previous works, most of them only show convergence from an initialization close to the true solution and thus require a carefully designed initialization scheme. 

5: However, random initialization has often been preferred by practitioners as it is model-agnostic.

6: In this paper, we show that ALS with random initialization converges to the true solution with $\varepsilon$-accuracy in $O(\log n + \log (1/\varepsilon)) $ iterations using only a near-optimal amount of samples, where we assume the measurement matrices to be i.i.d. Gaussian and where by $n$ we denote the ambient dimension.

7: Key to our proof is the observation that the trajectory of the ALS iterates only depends very mildly on certain entries of the random measurement matrices.

8: Numerical experiments corroborate our theoretical predictions.

9: \end{abstract}

10: