Tiny Emergence Machine

Control surface

Shape the conditions.

Rather than periodically retraining a fixed model, this miniature system updates continuously, learns the signal's local direction of change, and adjusts how quickly it reorganizes in response to prediction error.

Input signalGenerated

Upload CSV or text signal

Detects the most likely signal column, preferring named measures such as midpoint, price, close, value, or signal while ignoring index and time fields. Processing remains on this device.

Stream speed1.0×

Adaptive sensitivityBalanced

Skill vs. persistenceA signed forecast-skill score compares the adaptive model with a persistence baseline. Positive values mean the model is outperforming “repeat the last value”; negative values mean persistence is currently more accurate.

Adaptive plasticityThe current willingness to reorganize. It rises when error signals change and settles when the stream becomes predictable.

DriftHow strongly the signal’s local direction and pace are changing relative to its recent behavior. Higher values indicate a less stable trajectory.

SignalPrediction

Skill vs. persistence

—AVG —

Persistence advantageAdaptive skill advantage

Rolling forecast skill relative to a persistence baseline. Zero means equal error; positive values favor the adaptive model, while negative values favor repeating the previous observation. As a practical guide, 0–10 indicates only a slight advantage, 10–20 a modest advantage, 20–30 good skill, 30–40 strong skill, and values above 40 very strong skill. These ranges are interpretive rather than universal and should be considered alongside the signal, window length, and evaluation conditions.

Adaptive plasticity

—AVG —

How strongly the model is currently revising itself in response to drift and surprise.

Drift

—AVG —

How strongly the signal’s recent direction and pace are departing from its established local pattern.

Local landscapefast

Regional landscapeintermediate

Global landscapeslow

Raw MAE—Original signal units

Persistence MAE—Repeat-last-value baseline

Normalized MAE—Standardized signal scale

Skill window30Most recent samples

Samples 0Normalized error —Local browser process