Based on this observation, we adopt ageneralized version ofthe JensenShannon divergence for multiple distributions to encourage consistency around data points. Using this loss function, we show state-of-the-art results on both synthetic(CIFAR),andreal-world(e.g.WebVision)noisewithvaryingnoiserates.

It is not clear, however, whether these theoretical results apply to actual distributions such as images.

For`, we construct the systems of hyperrectangles by4 first precomputing anapproximate k-NN distance estimate using Ball Trees foreachdata point, andthen clustering5 the top-q densest data points intoT partitions using the k-means algorithm, where we binary search for the optimal6 parameterq.

The ability of machine learning to generate structured objects has progressed dramatically in the lastfewyears.

Compared meta-learning (2)results wecanonlymoutofM tasks.

For example, in image object recognition and language models building[4,5,6,7,8,9], the number of classes scales as the number of possible objects or the dictionary size respectively.