| 1. | Details for computing the inverse error function can be found at.
|
| 2. | The model is matched by the minimization of an error function.
|
| 3. | :* The Error function certainly has do with it.
|
| 4. | Therefore the error function satisfies the following recurrence relation:
|
| 5. | Exponential bounds and a pure exponential approximation for the complementary error function are given by
|
| 6. | This mimics closely how the error function in TD is used for reinforcement learning.
|
| 7. | Where \ textrm { erfc } ( z ) is the complementary error function.
|
| 8. | Where erfc is the complementary error function.
|
| 9. | However, the final potential function is not the 0-1 loss error function.
|
| 10. | The error function and its approximations can be used to estimate results that hold with high probability.
|
