Machine Learning / Evaluated demonstration
Combine physical simulation and clustering to explore motion patterns.
Actual recording · Explanatory English captions · Original synthetic data
Recorded demonstration of a functional local application. A pilot defines data, access and deployment; this website serves the video and screenshots.
THE PROBLEM
THE WORKFLOW
RK4 integrates a Duffing oscillator and extracts seven features. A model trained on 160 trajectories learns a PCA embedding and four clusters. Change amplitude, frequency, damping and initial position; animate the phase portrait and compare neighbours, spectrum and Poincare section.
RECORDED EVIDENCE
| Check | Cases |
|---|---|
| Documented checks; synthetic data | 30 of 30 |
30 trajectories with heldout seeds pass numerical and output checks. Training silhouette is 0.680; this is not accuracy. A short RK4 step comparison at 0.01 and 0.005 yields a maximum error of 6.67 × 10⁻¹⁰. Timings measure simulation and analysis on CPU, excluding HTTP and training.
WHAT WE COULD BUILD WITH YOU
Equations, units, parameter domain, solver, observables and numerical validation.
Simulated dimensionless system. Clusters are learned patterns, not physical labels or proof of chaos. The 30 checks verify integration and valid outputs; they do not measure regime-classification accuracy.
Scope a Machine Learning pilot ↗
Explore demo
Sensors → prediction → detected change
Separate changing inputs from declining model performance.
30 of 30 checks · Original synthetic dataDocumented checks; synthetic data
Explore demo
Trials → uncertainty → next design
Recommend trials using Bayesian optimisation and resource limits.
10 of 10 checks · Original synthetic dataDocumented checks; synthetic data