Mixture Representation Learning with Coupled Autoencoding Agents

Yeganeh Marghi, Rohan Gala, Uygar Sümbül

Keywords: Multi-agent network, representation learning, collective decision making, type-preserving data augmentation

Abstract: Jointly identifying a mixture of discrete and continuous factors of variability can help unravel complex phenomena. We study this problem by proposing an unsupervised framework called coupled mixture VAE (cpl-mixVAE), which utilizes multiple interacting autoencoding agents. The individual agents operate on augmented copies of training samples to learn mixture representations, while being encouraged to reach consensus on the categorical assignments. We provide theoretical justification to motivate the use of a multi-agent framework, and formulate it as a variational inference problem. We benchmark our approach on MNIST and dSprites, achieving state-of-the-art categorical assignments while preserving interpretability of the continuous factors. We then demonstrate the utility of this approach in jointly identifying cell types and type-specific, activity-regulated genes for a single-cell gene expression dataset profiling over 100 cortical neuron types.

One-sentence Summary: We propose a multi-agent variational framework to jointly infer discrete and continuous factors through collective decision making.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=vf_koE9GAV