PPolySim OS

AI Surrogate Model

Train a machine-learning surrogate on our real particle solver, then predict outcomes across the whole parameter space in microseconds — the instant-preview idea behind PhysicsX and Neural Concept, running in your browser.

AI Surrogate StudioLive

Controls

Train a surrogate model on our real particle solver, then get near-instant predictions instead of re-running the full simulation — the same idea behind PhysicsX and Neural Concept, running in your browser.

Presets

▶ Run in Python

Data Inspector

Train the surrogate to see accuracy metrics.

Governing equation

Runs locally in your browser — free forever. Scale to the cloud when reality gets heavy.

or unlock everything with Pro →
★ Sign in to save this setup
Save your tuned setup, or drop this simulation into your own site, docs, or course page.

How it works

We sample the real solver across a grid of parameters, then fit a Gaussian radial-basis-function (RBF) interpolator with a linear tail and ridge regularization — solving the interpolation system directly. The trained surrogate predicts new outputs by evaluating basis functions, which is orders of magnitude faster than re-simulating. Accuracy is reported honestly using a held-out test grid (R², RMSE), and you can verify any prediction against the full solver to see the real speed-up.

Ask the AI about this model

The math, the assumptions, real-world uses, or a code translation — explained for this exact simulation.

Related live simulations

Frequently asked questions

Is this AI surrogate model tool really free?
Yes. AI Surrogate Model runs entirely in your browser using your device's own compute, so local use is free forever. You only pay Compute Tokens if you scale a job to the cloud.
Do I need to install anything?
No. Everything runs client-side in a modern browser — no downloads, no license, no account required to start.
Can I save or share my simulation?
Create a free account to save projects, and use a shareable embed or minted DOI to publish a live, interactive version anywhere.
How accurate are the results?
The solver uses established numerical methods, but results are for research and educational purposes and should be validated against experiment or professional review before you rely on them.