Gravitational Wave Detection

Gravitational Wave Detection

Join the Competition Contribute to Benchmark

Standard reconstruction benchmark — forward model perfectly known, no calibration needed. Score = 0.5 × clip((PSNR−15)/30, 0, 1) + 0.5 × SSIM

# Method Score PSNR (dB) SSIM Source
🥇 WaveFormer 0.706 30.5 0.895 ✓ Certified GW detection transformer, 2024
🥈 GW-CNN 0.655 28.8 0.850 ✓ Certified George & Huerta, Phys. Rev. D 2018
🥉 BayesWave 0.513 24.5 0.710 ✓ Certified Cornish & Littenberg, CQG 2015
4 Matched Filter 0.343 20.0 0.520 ✓ Certified Allen et al., Phys. Rev. D 2012

Dataset: PWM Benchmark (4 algorithms)

Blind Reconstruction Challenge — forward model has unknown mismatch, must calibrate from data. Score = 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

# Method Overall Score Public
PSNR / SSIM
 Dev
PSNR / SSIM
 Hidden
PSNR / SSIM
 Trust Source
🥇 WaveFormer + gradient 0.657
0.736
28.88 dB / 0.898
0.651
25.92 dB / 0.830
0.583
23.33 dB / 0.744
✓ Certified GW detection transformer, 2024
🥈 GW-CNN + gradient 0.603
0.686
26.92 dB / 0.856
0.593
23.17 dB / 0.738
0.531
20.66 dB / 0.630
✓ Certified George & Huerta, Phys. Rev. D 2018
🥉 BayesWave + gradient 0.526
0.611
23.01 dB / 0.731
0.535
20.87 dB / 0.640
0.432
17.41 dB / 0.471
✓ Certified Cornish & Littenberg, CQG 2015
4 Matched Filter + gradient 0.430
0.434
17.06 dB / 0.453
0.433
17.22 dB / 0.461
0.423
17.56 dB / 0.478
✓ Certified Allen et al., Phys. Rev. D 2012

Complete score requires all 3 tiers (Public + Dev + Hidden).

Join the competition →
Scoring: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖) PSNR 40% · SSIM 40% · Consistency 20%
Public 5 scenes

Full-access development tier with all data visible.

What you get & how to use

What you get: Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.

How to use: Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.

What to submit: Reconstructed signals (x_hat) and corrected spec as HDF5.

Public Leaderboard
# Method Score PSNR SSIM
1 WaveFormer + gradient 0.736 28.88 0.898
2 GW-CNN + gradient 0.686 26.92 0.856
3 BayesWave + gradient 0.611 23.01 0.731
4 Matched Filter + gradient 0.434 17.06 0.453
Spec Ranges (4 parameters)
Parameter Min Max Unit
calibration_amplitude 0.99 1.02 -
phase_calibration -0.01 0.02 rad
power_spectral_density -0.15 0.15 1/Hz
timing_offset -0.15 0.15 s
View Details →
Dev 5 scenes

Blind evaluation tier — no ground truth available.

What you get & how to use

What you get: Measurements (y), ideal forward operator (H), and spec ranges only.

How to use: Apply your pipeline from the Public tier. Use consistency as self-check.

What to submit: Reconstructed signals and corrected spec. Scored server-side.

Dev Leaderboard
# Method Score PSNR SSIM
1 WaveFormer + gradient 0.651 25.92 0.83
2 GW-CNN + gradient 0.593 23.17 0.738
3 BayesWave + gradient 0.535 20.87 0.64
4 Matched Filter + gradient 0.433 17.22 0.461
Spec Ranges (4 parameters)
Parameter Min Max Unit
calibration_amplitude 0.988 1.018 -
phase_calibration -0.012 0.018 rad
power_spectral_density -0.15 0.15 1/Hz
timing_offset -0.15 0.15 s
Submit Reconstruction View Details →
Hidden 5 scenes

Fully blind server-side evaluation — no data download.

What you get & how to use

What you get: No data downloadable. Algorithm runs server-side on hidden measurements.

How to use: Package algorithm as Docker container / Python script. Submit via link.

What to submit: Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

Hidden Leaderboard
# Method Score PSNR SSIM
1 WaveFormer + gradient 0.583 23.33 0.744
2 GW-CNN + gradient 0.531 20.66 0.63
3 BayesWave + gradient 0.432 17.41 0.471
4 Matched Filter + gradient 0.423 17.56 0.478
Spec Ranges (4 parameters)
Parameter Min Max Unit
calibration_amplitude 0.993 1.023 -
phase_calibration -0.007 0.023 rad
power_spectral_density -0.15 0.15 1/Hz
timing_offset -0.15 0.15 s
Submit Algorithm View Details →

Blind Reconstruction Challenge

Challenge

Given measurements with unknown mismatch and spec ranges (not exact params), reconstruct the original signal. A method must be evaluated on all three tiers for a complete score. Scored on a composite metric: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖).

Input

Measurements y, ideal forward model H, spec ranges

Output

Reconstructed signal x̂

Spec DAG — Forward Model Pipeline

P → Σ → D

P Propagation
Σ Summation
D Detector

Mismatch Parameters

Symbol Parameter Description Nominal Perturbed
c_a calibration_amplitude Calibration amplitude (-) 1.0 1.01
p_c phase_calibration Phase calibration (rad) 0.0 0.01
p_s power_spectral_density Power spectral density (1/Hz) 1e-23 0.0
t_o timing_offset Timing offset (s) 0.0 0.0

Credits System

40%
Platform Profit Pool
Revenue allocated to benchmark rewards
30%
Winner Share
Top algorithm receives from pool
$100
Min Withdrawal
Minimum payout threshold
Spec Primitives Reference (11 primitives)
P Propagation

Free-space or medium propagation kernel (Fresnel, Rayleigh-Sommerfeld).

M Mask / Modulation

Spatial or spatio-temporal amplitude modulation (coded aperture, SLM pattern).

Π Projection

Geometric projection operator (Radon transform, fan-beam, cone-beam).

F Fourier Sampling

Sampling in the Fourier / k-space domain (MRI, ptychography).

C Convolution

Shift-invariant convolution with a point-spread function (PSF).

Σ Summation / Integration

Summation along a physical dimension (spectral, temporal, angular).

D Detector

Sensor readout with gain g and noise model η (Gaussian, Poisson, mixed).

S Structured Illumination

Patterned illumination (block, Hadamard, random) applied to the scene.

W Wavelength Dispersion

Spectral dispersion element (prism, grating) with shift α and aperture a.

R Rotation / Motion

Sample or gantry rotation (CT, electron tomography).

Λ Wavelength Selection

Spectral filter or monochromator selecting a wavelength band.