You are not logged in. Your edit will be placed in a queue until it is peer reviewed.
We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.
-
1$\begingroup$ How hard is it to rename the variables? $\endgroup$– Andrej BauerCommented Dec 5, 2023 at 7:56
-
$\begingroup$ “queries for neural network solvers may only reference a fixed, finite set of variables”. What does this mean? What “neural network solvers” are you talking about? $\endgroup$– Jason RuteCommented Dec 5, 2023 at 14:22
-
$\begingroup$ So you have one variable for every input and output value for the neural network, i.e. if your neural network has 3 inputs and 2 outputs, your query can only refer to 5 variables in total. It's impossible to rename these variables. For some solvers (e.g. Marabou) this is an artificial limitation, but for most solvers (e.g. alpha-beta-crown) it is a pretty hard limitation. $\endgroup$– Matthew DaggittCommented Dec 5, 2023 at 14:54
-
1$\begingroup$ I think you and I have a very different definition of what a “neural network solver” means. For me it is a NN-based tool for solving theorems. For you I think it is a SAT-based tool for verifying neural network properties, right? If renaming the variables like Andre suggested isn’t an option, then I think you need to explain more about the domain and how this limitation comes up. $\endgroup$– Jason RuteCommented Dec 5, 2023 at 15:02
-
$\begingroup$ Yes, they're essentially SAT-solver like tools for verifying properties of neural networks. I'm not really keen to go into the specifics of why they have these limitations. I was just wondering if anyone was aware of any other domains where solvers with fixed, finite sets of variables have been integrated in proof assistants? $\endgroup$– Matthew DaggittCommented Dec 5, 2023 at 15:21
|
Show 2 more comments
How to Edit
- Correct minor typos or mistakes
- Clarify meaning without changing it
- Add related resources or links
- Always respect the author’s intent
- Don’t use edits to reply to the author
How to Format
-
create code fences with backticks ` or tildes ~
```
like so
``` -
add language identifier to highlight code
```python
def function(foo):
print(foo)
``` - put returns between paragraphs
- for linebreak add 2 spaces at end
- _italic_ or **bold**
- indent code by 4 spaces
- backtick escapes
`like _so_`
- quote by placing > at start of line
- to make links (use https whenever possible)
<https://example.com>
[example](https://example.com)
<a href="https://example.com">example</a> - MathJax equations
$\sin^2 \theta$
How to Tag
A tag is a keyword or label that categorizes your question with other, similar questions. Choose one or more (up to 5) tags that will help answerers to find and interpret your question.
- complete the sentence: my question is about...
- use tags that describe things or concepts that are essential, not incidental to your question
- favor using existing popular tags
- read the descriptions that appear below the tag
If your question is primarily about a topic for which you can't find a tag:
- combine multiple words into single-words with hyphens (e.g. type-theory), up to a maximum of 35 characters
- creating new tags is a privilege; if you can't yet create a tag you need, then post this question without it, then ask the community to create it for you