Fuse It or Lose It: Deep Fusion for Multimodal Simulation-Based Inference
This work addresses the challenge of multimodal data integration for researchers in fields like neuroscience and cardiology, offering incremental improvements through novel fusion strategies.
The authors tackled the problem of integrating heterogeneous data from different sources in simulation-based inference by proposing MultiNPE, a method that uses deep fusion approaches, and found that late and hybrid fusion techniques outperformed single-source baselines and improved inference accuracy in scientific models from cognitive neuroscience and cardiology.
We present multimodal neural posterior estimation (MultiNPE), a method to integrate heterogeneous data from different sources in simulation-based inference with neural networks. Inspired by advances in deep fusion, it allows researchers to analyze data from different domains and infer the parameters of complex mathematical models with increased accuracy. We consider three fusion approaches for MultiNPE (early, late, hybrid) and evaluate their performance in three challenging experiments. MultiNPE not only outperforms single-source baselines on a reference task, but also achieves superior inference on scientific models from cognitive neuroscience and cardiology. We systematically investigate the impact of partially missing data on the different fusion strategies. Across our experiments, late and hybrid fusion techniques emerge as the methods of choice for practical applications of multimodal simulation-based inference.