Skip to content
Home / Catalog / Infer a Size-Dependent Ag Dielectric Model from Paired-Strip Spectra

Infer a Size-Dependent Ag Dielectric Model from Paired-Strip Spectra

  • Task ID: physics.ag_strip_dielectric_inference
  • Domain: physics
  • Subdomain: nanophotonics
  • Status: final
  • Benchmark set: seed42 (60 tasks)
  • Tags: metamaterials, plasmonics, silver, dielectric_function, model_fitting, spectra, inverse_problem, hypothesis_testing

Runtime and requirements

  • Estimated time: 45-120 minutes
  • Python: >=3.11
  • Packages: numpy>=1.26, pandas>=2.0, matplotlib>=3.8, scipy>=1.11
  • GPU required: no
  • Network required: no

Public input and output contract

Inputs

  • calibration_samples.csv (data): Sample metadata for the calibration set, including strip geometry, roughness, polarization, and train/heldout split.
  • calibration_spectra.csv (data): Synthetic transmission/reflection/optical-density spectra for the calibration samples.
  • heldout_samples.csv (data): Metadata for multiple unseen paired-strip samples spanning different transfer regimes.
  • modeling_notes.md (text): Definitions of the surrogate dielectric model, observables, and expected units.

Outputs

  • analysis.py (code): Agent-authored fitting and prediction script.
  • results/fitted_params.json (data): Recovered dielectric-model parameters with keys including preferred_model, alpha_tm, alpha_te, roughness_tm, roughness_te, bulk_tm, and bulk_te.
  • results/model_comparison.csv (data): One row per candidate model family with calibration_rmse, crossval_rmse, num_parameters, preferred flag, and a short mechanism summary.
  • results/heldout_predictions.csv (data): Predicted held-out spectra versus wavelength for all held-out samples and both polarizations, generated by the same selected compact model family reported in model_comparison.csv.
  • results/resonance_summary.csv (data): Calibration and held-out resonance metrics with numeric short-band and long-band peak summaries plus off-resonance optical density; bandwise maxima should be reported even when one mode is weak.
  • results/diagnostic_plots.png (figure): Figure summarizing calibration fits, multi-heldout predictions, and inferred loss trends.
  • results/conclusion.txt (conclusion): Short scientific conclusion stating which mechanism is supported by the data and why.

Public repository files

View this task in ASI-Bench

Formal benchmark task directories in the public repository are metadata-only. Versioned prompts and inputs are distributed through the pinned benchmark dataset.

Provenance

This task is included in the current seed42 benchmark set at dataset revision f11a199f71fb4b854ac43a1bf548d5df519141a9.

This page is generated only from files tracked in the public ASI-Bench repository at commit f18382f03faf. Submission bundles, run logs, private scoring configuration, and private reference answers are not read by this page generator.