1
Fork 0
mirror of https://github.com/Steffo99/appunti-magistrali.git synced 2024-11-26 03:54:17 +00:00
appunti-steffo/7 - Fondamenti di machine learning/2 - Calcolo vettoriale/derivata direzionale unilaterale.md

54 lines
852 B
Markdown
Raw Normal View History

2024-08-23 11:13:05 +00:00
Concetto di [[derivata]] applicato ai [[array|vettori]].
$$
\Huge
2024-11-06 12:42:11 +00:00
\derivativeDirectionalOf{x}{s}
2024-08-23 11:13:05 +00:00
$$
È implicato che sia sempre la derivata positiva:
$$
\Large
\Huge
2024-11-06 12:42:11 +00:00
\derivativeDirectionalOf
[\color{lime} \mathrm{f}]
{\color{cyan} x}
{\color{orange} +s}
2024-08-23 11:13:05 +00:00
=
2024-11-06 12:42:11 +00:00
\lim_{\smol \to 0^{\color{orange} +}}
\frac
{
{\color{lime} \mathrm{f}}
\par{
{\color{cyan} x} + \smol \cdot {\color{orange} s}
}
-
{\color{lime} \mathrm{f}}
\par{
{\color{cyan} x}
}
2024-08-23 11:13:05 +00:00
}{
\smol
}
$$
La derivata negativa è invece definita come:
$$
\Large
2024-11-06 12:42:11 +00:00
\derivativeDirectionalOf[\color{lime} -f]{{\color{cyan} x_0}}{{\color{orange} -s}}
2024-08-23 11:13:05 +00:00
=
2024-11-06 12:42:11 +00:00
\lim_{\smol \to 0^{\color{orange} -}}
\frac
{
{\color{lime} \mathrm{f}}
\par{
{\color{cyan} x} + \smol \cdot {\color{orange} s}
}
-
{\color{lime} \mathrm{f}}
\par{
{\color{cyan} x}
}
2024-08-23 11:13:05 +00:00
}{
\smol
}
$$