Prototypical Networks for Few-shot Learning — mentioned by Eridrus

Mon, 12 Mar 2018 16:48:23 GMT

Ok, I admit I hadn't looked up the actual numbers before I posted that comment, but the Snell paper wasn't the last word on that line of work, it was followed by https://arxiv.org/abs/1703.05175 (which doesn't have miniImagenet results). And there may have been further work in that direction that I'm not aware of. You're right that Reptile is the simplest recent algorithm in the meta-learning literature, but I would argue that's basically my point, they started from somewhere pretty ambitious (lets learn a learner, or at least an SGD update rule), and ended up with learning an initialization…

Paper and discussion history. A mention is not an endorsement.

Prototypical Networks for Few-shot Learning · HackerPapers discussions

Prototypical Networks for Few-shot Learning — mentioned by Eridrus