@Article{cmc.2026.082976,
AUTHOR = {Mohammed Hassan Alnemari, Abdelrahman Osman Elfaki, Anas Bushnag, Mohamed Hussien Mohamed Nerma},
TITLE = {Conformal Prediction for Reliable Hyperparameter Selection in Sparse FIR Filter Design},
JOURNAL = {Computers, Materials \& Continua},
VOLUME = {},
YEAR = {},
NUMBER = {},
PAGES = {{pages}},
URL = {http://www.techscience.com/cmc/online/detail/27246},
ISSN = {1546-2226},
ABSTRACT = {Sparse finite impulse response (FIR) filters reduce computational cost on resource-constrained devices, but selecting the sparsification threshold <mml:math id="mml-ieqn-1"><mml:mi>λ</mml:mi></mml:math> is typically left to grid search or hand tuning. We propose a two-stage method: a 67,331-parameter surrogate network predicts <mml:math id="mml-ieqn-2"><mml:mo stretchy="false">(</mml:mo><mml:msub><mml:mi>A</mml:mi><mml:mi>p</mml:mi></mml:msub><mml:mo>,</mml:mo><mml:msub><mml:mi>A</mml:mi><mml:mi>s</mml:mi></mml:msub><mml:mo>,</mml:mo><mml:mi>S</mml:mi><mml:mo stretchy="false">)</mml:mo></mml:math> (passband ripple in dB, stopband attenuation in dB, sparsity in %) from a filter specification and a candidate <mml:math id="mml-ieqn-3"><mml:mi>λ</mml:mi></mml:math>, and split conformal prediction (CP) calibrates ± intervals around each prediction. We then select <mml:math id="mml-ieqn-4"><mml:mi>λ</mml:mi></mml:math> by minimizing a worst-case penalty computed on the conservative ends of the intervals (the upper bound on <mml:math id="mml-ieqn-5"><mml:msub><mml:mi>A</mml:mi><mml:mi>p</mml:mi></mml:msub></mml:math> and the lower bound on <mml:math id="mml-ieqn-6"><mml:msub><mml:mi>A</mml:mi><mml:mi>s</mml:mi></mml:msub></mml:math>). On 10,000 test specifications the method reaches 76.5% specification satisfaction, near-parity with grid search (78.4%) with a 1.9<mml:math id="mml-ieqn-7"><mml:mo>×</mml:mo></mml:math> speedup, while point-prediction surrogates reach only 39.4%. On feasible specifications (where any grid <mml:math id="mml-ieqn-8"><mml:mi>λ</mml:mi></mml:math> satisfies both constraints), the method reaches 97.6%. Stratified (Mondrian) conformal prediction lifts standard CP coverage from 67%–75% to 95.5%, and adaptive recalibration brings passband coverage to 91.3%. The procedure transfers without modification to iteratively reweighted least squares (IRLS) sparsification (76.6%) and to highpass (79.2%) and bandpass (52.4%) filters. The implementation runs on a central processing unit (CPU) and is suitable for edge deployment; code and data are public.},
DOI = {10.32604/cmc.2026.082976}
}