Read the equation
Rewards utility while penalizing the magnitude of the change signal. Both contributions use the same coefficient, so the increment is 0.05(u−|dx|).
Its role in the build
Rewards utility while penalizing the magnitude of the change signal.
Terms & definitions
| Term | Meaning in this source |
|---|---|
| x | current scalar family state |
| dx | change signal |
| d_x / dx_norm | normalized change signal |
| u / utility | utility input |
| e / err | error input |
| s / stab | stability input |
| φ | golden-ratio constant |
How it works
Read the displayed expression together with the surrounding source statement; it is retained as an expression rather than silently completed into an equality.
Substitute the source-defined variables or recorded values and follow the displayed arithmetic and indexing exactly.
Use this version’s variable definitions and update order. The same symbol may be assigned a different meaning in another build.
Assumptions & limits
Conditions
The surrounding implementation must define the input scales and any output clipping.
The source does not define a probability model or variational bound that would identify this with variational free energy.
Original context
Exact original notation
x - 0.05|dx| + 0.05×u
Related builds
Sources & provenance
x - 0.05|dx| + 0.05×u