Compact neural networks for astronomy with optimal transport bias correction
This addresses the problem of limited large-scale morphological classification and redshift prediction in astronomy, representing a domain-specific advancement with interdisciplinary impact.
The paper tackles the efficiency-resolution tradeoff in astronomical imaging by introducing WaveletMamba, a framework that achieves 81.72% classification accuracy at 64x64 resolution with only 3.54M parameters and 9.7x computational efficiency gains, while also improving bias correction by 22.96% in Log-MSE.
Astronomical imaging confronts an efficiency-resolution tradeoff that limits large-scale morphological classification and redshift prediction. We introduce WaveletMamba, a theory-driven framework integrating wavelet decomposition with state-space modeling, mathematical regularization, and multi-level bias correction. WaveletMamba achieves 81.72% +/- 0.53% classification accuracy at 64x64 resolution with only 3.54M parameters, delivering high-resolution performance (80.93% +/- 0.27% at 244x244) at low-resolution inputs with 9.7x computational efficiency gains. The framework exhibits Resolution Multistability, where models trained on low-resolution data achieve consistent accuracy across different input scales despite divergent internal representations. The framework's multi-level bias correction synergizes HK distance (distribution-level optimal transport) with Color-Aware Weighting (sample-level fine-tuning), achieving 22.96% Log-MSE improvement and 26.10% outlier reduction without explicit selection function modeling. Here, we show that mathematical rigor enables unprecedented efficiency and comprehensive bias correction in scientific AI, bridging computer vision and astrophysics to revolutionize interdisciplinary scientific discovery.