A better permutation-equivariance property for set prediction.

[arxiv] [code] [video] [summary] [poster]

To predict a set from a vector, use gradient descent to find a set the encodes to that vector.

Sort in encoder and undo sorting in decoder to avoid responsibility problem in set auto-encoders.

Learn how to permute a set, then encode permuted set with RNN to obtain a set representation.

Enabling visual question answering models to count by handling overlapping object proposals.