diff options
| author | ericmarin <maarin.eric@gmail.com> | 2026-06-19 12:39:13 +0200 |
|---|---|---|
| committer | ericmarin <maarin.eric@gmail.com> | 2026-06-26 09:57:03 +0200 |
| commit | d3e761a2286d04a3c0005b199653df2f6501f070 (patch) | |
| tree | bc4c77a68d94662ad41c67710e07af851e5d3287 /chapters/01-introduction.tex | |
| parent | 8eb2ce59ae307984a5b40a05cefec1f7e112fb02 (diff) | |
| download | vein-d3e761a2286d04a3c0005b199653df2f6501f070.tar.gz vein-d3e761a2286d04a3c0005b199653df2f6501f070.zip | |
refined core
Diffstat (limited to 'chapters/01-introduction.tex')
| -rw-r--r-- | chapters/01-introduction.tex | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/chapters/01-introduction.tex b/chapters/01-introduction.tex index fc4017a..8f46660 100644 --- a/chapters/01-introduction.tex +++ b/chapters/01-introduction.tex @@ -20,7 +20,7 @@ either hold or are violated. The kind of properties that can be verified are two An example of a Multi-Network property is equivalence between different neural networks; this verification is crucial when replacing a neural network with a smaller, equivalent one. Techniques based on specialized \textit{Satisfiability Modulo Theories} (SMT) solvers may be used in the verification -algorithm, however, as networks grow in depth and width, the number of non-linear components, most +algorithm. However, as networks grow in depth and width, the number of non-linear components, most notably the \textit{Rectified Linear Unit}\footnote{Defined as $f(x)=\max(0, x)$, where $x$ is the input to the neuron.} (ReLU) activation function, creates an exponential search space for the solver. Because of this issue, running a solver against a raw, unoptimized network is often computationally prohibitive. |
