2050e03ca119580f74cca14cc6e97462-Reviews.html

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. The authors propose sequential and active formulations for low rank matrix completion including a formulation where the underlying matrix/tensor is known to be noisy. The resultant formulations have strong performance bounds with regard to the number of entries that need to be polled as a function of the matrix rank. Theorem 3 is in my opinion a main contribution and it was good to see it was experimentally verified. The lower bound for uniform sampling given in theorem 4 is also useful, but it appears you did not experimentally verify how close your work is to this lower bound.