Search papers, labs, and topics across Lattice.
The paper introduces a sparse Bayesian functional deep neural network (sBayFDNN) to address the limitations of both linear functional models and black-box deep learning in analyzing complex, continuously structured data. sBayFDNN uses a deep Bayesian architecture to learn adaptive functional embeddings and a structured prior for interpretable, region-wise selection of influential domains with uncertainty quantification. The authors establish approximation error bounds, posterior consistency, and region selection consistency, and demonstrate sBayFDNN's superior performance in simulations and real-world applications, particularly in identifying functionally meaningful regions.
Finally, a deep learning model for functional data analysis with provable guarantees on region selection consistency.
In modern applications such as ECG monitoring, neuroimaging, wearable sensing, and industrial equipment diagnostics, complex and continuously structured data are ubiquitous, presenting both challenges and opportunities for functional data analysis. However, existing methods face a critical trade-off: conventional functional models are limited by linearity, whereas deep learning approaches lack interpretable region selection for sparse effects. To bridge these gaps, we propose a sparse Bayesian functional deep neural network (sBayFDNN). It learns adaptive functional embeddings through a deep Bayesian architecture to capture complex nonlinear relationships, while a structured prior enables interpretable, region-wise selection of influential domains with quantified uncertainty. Theoretically, we establish rigorous approximation error bounds, posterior consistency, and region selection consistency. These results provide the first theoretical guarantees for a Bayesian deep functional model, ensuring its reliability and statistical rigor. Empirically, comprehensive simulations and real-world studies confirm the effectiveness and superiority of sBayFDNN. Crucially, sBayFDNN excels in recognizing intricate dependencies for accurate predictions and more precisely identifies functionally meaningful regions, capabilities fundamentally beyond existing approaches.
