Review for NeurIPS paper: Bootstrap Your Own Latent - A New Approach to Self-Supervised Learning

Weaknesses: -As mentioned in the paper, the proposed method has a trivial solution, that both models output 0's. To me, the method is too simple to be true. I tried to reimplement it, but no success. It is highly recommend to opensource the code for reproduceable research. How can you learn detection with frozen representation? Please use the standard settings, e.g. as in MoCo.