First use of the learned harmonic mean estimator for Bayesian model selection across circular/eccentric, white-noise/GP, and trend variants in radial velocity exoplanet analyses.
Mixed citations
A new look at the statistical model identification
Mixed citation behavior. Most common role is background (58%).
39
Pith papers citing it
Background
58% of classified citations
citation-role summary
background 8
method 4
citation-polarity summary
representative citing papers
Time-resolved XRISM spectroscopy of NGC 4395 reveals variable inner-disk Fe Kα emission interpreted as Lense-Thirring precession, favoring low black hole mass (~9e3 solar masses) and moderate spin (a≳0.6).
CMS measures Υ(1S,2S,3S) → τ⁺τ⁻ with 5.8σ significance and cross section 3.5 ± 0.7 (stat) ± 0.7 (syst) nb while setting 95% CL upper limits of 40-400 pb on spin-zero resonance production to ττ with no observed excess in 61.9 fb⁻¹ of 13.6 TeV data.
Human face perception aligns with neural networks trained on inverse-generative and naturalistic discriminative tasks, as these best predict human dissimilarity judgments on controversial and random face pairs.
The mHz QPO in 1ES 1927+654 has persisted and stabilized at ~2.5 mHz with stable soft lags, no strong harmonic, and repeatable large-amplitude flux patterns, also seen in NuSTAR.
Protein length CCDFs in 22 organisms yield Tsallis q values that separate into superextensive, boundary, and subextensive regimes aligned with the bilaterian transition, with q rising monotonically in deuterostomes.
Kolmogorov n-width theory plus PRESS statistics yield closed-form optimal spline resolution; KORE estimates bias/noise scales from two pilots and matches CV performance with far fewer fits.
Predictively consistent priors let complex Bayesian models match or beat the out-of-sample performance of selected simpler models across linear, logistic, and nonlinear examples without explicit selection.
Multi-wavelength MCMC modeling of RCW 38 supports hadronic gamma-ray production with K_ep ≲ 10^{-3} and acceleration efficiency ≳1%, consistent with cosmic-ray composition requirements.
A QCD-vacuum-based model of dynamical dark energy fits Planck+ACT+SPT, DESI DR2, and supernova data while reproducing the late-time evolution favored by DESI.
Proves finite-shot mean-squared-error laws for virtual distillation and symmetry verification that define certified operating windows and a selection trichotomy for their comparison.
Derives exact operating characteristic corrections and a numerical search over sample sizes to obtain optimal two-stage Bayes factor designs for two-arm binary-endpoint phase II trials that minimize expected sample size under the null.
Bitcoin price power law is rejected on distributional series and not robust to time-origin shifts, but dominates medium-horizon forecasts against baselines because it avoids committing to specific wave shapes.
Proposes adaptive multiple importance sampling for robust Bayesian model evidence estimation under parameter non-identifiability, shown to outperform deterministic methods on ecological case studies while being cheaper than MCMC.
Discovery of extended GeV gamma-ray emission from AFGL 490, with the protostellar jet identified as the likely particle accelerator rather than stellar winds.
HONKAI identifies 193 dense cores in 16 clumps within three IRDCs, finding most have virial ratios >1 but mass-size relations below the massive star formation threshold and a steeper high-mass CMF slope.
A framework for cross-validation optimal feature selection in linear SVM classification is developed by reformulating the bilevel problem into a single-level mixed-integer optimization using LS-SVM, with simulation results indicating competitive performance.
DESI data on the GD-1 stream identifies a thin cold core and a hot cocoon with 30% of members whose dispersion is consistent with 11 Gyr of dark matter subhalo heating.
Three complementary metrics based on statistical heterogeneity and Minkowski functionals enable unsupervised selection of defect-representative images from IRT sequences without prior spatial knowledge.
XRISM spectroscopy of NGC 4388 shows the Fe Kα line arising from the torus, inner torus edge, and BLR at distinct radii, with absorption revealing a gravitationally bound failed wind.
Updated M_BH-R_b relation for core-Sérsic galaxies has slope 1.16 and 0.28 dex scatter; large-core galaxies drive a high-mass upturn in M_BH-σ attributed to successive dry mergers.
Cultural zones explain variance in safety ratings beyond demographics across six datasets, with roughly 10% of items identified as culturally sensitive.
Post-selection with DL or FBF after multi-objective GP search improves test-set performance over AIC/BIC baselines on noisy synthetic and real regression tasks, while using DL directly as fitness often causes premature convergence to overly simple models.
New JWST spectrum of a z=8.3 galaxy with prior ALMA dust detection shows AGN signatures, 0.15 Zsun metallicity, Mdust ~10^6 Msun, Tdust~91 K, and an [OIII] line ratio higher than nebular models allow.
citing papers explorer
-
To select or not to select: predictively consistent priors instead of model selection
Predictively consistent priors let complex Bayesian models match or beat the out-of-sample performance of selected simpler models across linear, logistic, and nonlinear examples without explicit selection.
-
Optimal sequential two-stage Bayes Factor Design for two-arm clinical Phase II Trials with binary Endpoints
Derives exact operating characteristic corrections and a numerical search over sample sizes to obtain optimal two-stage Bayes factor designs for two-arm binary-endpoint phase II trials that minimize expected sample size under the null.
-
Reliable model selection in the presence of parameter non-identifiability
Proposes adaptive multiple importance sampling for robust Bayesian model evidence estimation under parameter non-identifiability, shown to outperform deterministic methods on ecological case studies while being cheaper than MCMC.