Sum-Product Autoencoding: Encoding and Decoding Representations using Sum-Product Networks

Sum-Product Networks (SPNs) are a deep probabilistic architecture that up to now has been successfully employed for tractable inference. Here, we extend their scope towards unsupervised representation learning: we encode samples into continuous and categorical embeddings and show that they can also be decoded back into the original input space by leveraging MPE inference. We characterize when this Sum- Product Autoencoding (SPAE) leads to equivalent recon- structions and extend it towards dealing with missing embed- ding information. Our experimental results on several multilabel classification problems demonstrate that SPAE is com- petitive with state-of-the-art autoencoder architectures, even if the SPNs were never trained to reconstruct their inputs.