Hello all!

So I am working on something that requires me to compare the runtime/operational cost (in flops) for the least squares problem. The 3 general purpose methods I am aware of are :

1. Normal Equations by Cholesky Factorization

2. QR Factorization

3. SVD

ref : http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.386.5345&rep=rep1&type=pdf

I was curious about any recent work that improves upon the time complexity for computing least squares.

So far what I found is a meta-heuristic method :

ref : Article Solving Linear Least Squares Problems Based on Improved Cuck...

and a randomized algorithm.

ref: https://arxiv.org/pdf/0710.1435

But I wanted to ask the community (who is more experienced and well-versed with the literature) are there some algorithms I should be aware of (or some updates till date that helped make them faster)?