Hellinger Multimodal Variational Autoencoders

1. 1Revisit multimodal inference using probabilistic opinion pooling
2. 2Apply Hellinger distance to aggregate unimodal inference distributions
3. 3Approximate the joint posterior using the proposed method
4. 4Evaluate the performance of the Hellinger Multimodal VAEs on multimodal datasets

ML researchers and engineers working on multimodal generative models can benefit from this approach to improve inference and learning

💡 Probabilistic opinion pooling can be used to improve multimodal inference in VAEs

🤖 Hellinger Multimodal VAEs revise multimodal inference using probabilistic opinion pooling!