{"id":"arxiv/2610.17023","url":"https://arxiv.org/abs/2610.17023","tags":["math.NA","math.OC"],"score":0,"title":"Edge-preserving Bayesian neural networks via hierarchical sparsity-promoting priors","domain":"math.NA","source":"arxiv","authors":["Jan Habscheid","Jan Glaubitz"],"reviews":0,"subject":"math.NA","summary":"Recovering piecewise-smooth functions with jump discontinuities from noisy, indirect, and incomplete data is a central task in computational inverse problems. Bayesian neural networks (BNNs) with heavy-tailed weight priors have emerged as effective edge-preserving priors for such problems. However, posterior inference for these models remains computationally challenging, and existing approaches settle for point estimates obtained by directly optimizing heavy-tailed log-densities. We propose GSBL-BNNs, which model the neural network parameters hierarchically using sparsity-promoting priors…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-12","reactions":{}}