This is a really interesting way of putting it. Is the presence of e.g. 31/32 of the original data the only reason InstaHide can train a model at all? I haven’t

This is a really interesting way of putting it. Is the presence of e.g. 31/32 of the original data the only reason InstaHide can train a model at all? I haven’t read the original paper, but it seems very possible that the entire approach runs into a fundamental tradeoff.