Skip to content
Snippets Groups Projects
Commit 9a96f475 authored by Augustin Lemesle's avatar Augustin Lemesle
Browse files

add acks, remove fake news, fix css and typos

parent 941d56de
No related branches found
No related tags found
No related merge requests found
......@@ -53,7 +53,7 @@
PyRAT stands for Python Reachability Assessment Tool and is based on abstract interpretation techniques with the aim of determining whether in a given neural network a certain state can be reach by the network from some initial parameters.
</p>
<p>
To do so, it propagates different types of abstract domains along the different layers of the neural network up until the output layer. This technique will overapproximate some of the operation of the layer and thus the output space reached will always be bigger than the exact output space. If the output space reached is inside a certain space defining the property we want, then we can say that the property is verified. Otherwise, we cannot conclude.
To do so, it propagates different types of abstract domains along the different layers of the neural network up until the output layer. This approach will overapproximate some of the operations of the layers and thus the output space reached will always be bigger than the exact output space. If the output space reached is inside a certain space defining the property we want, then we can say that the property is verified. Otherwise, we cannot conclude.
</p>
</div>
......@@ -159,7 +159,7 @@
<div class="column is-two-third">
<h2 class="subtitle">Input partitioning</h2>
<p>
Another key aspect of PyRAT lies in the input partitioning mechanisms that it employs. For lower dimensionality inputs, PyRAT is able to use input partitioning as a mean to increase the precision of the analysis and prove a wider range of property. This partitioning is reinforced by heuristics tailored to the abstract domains we use which allow a significant boost in precision on such networks<br>
Another key aspect of PyRAT lies in the input partitioning mechanisms that it employs. For lower dimensionality inputs, PyRAT is able to use input partitioning as a mean to increase the precision of the analysis and prove a wider range of property. This partitioning is reinforced by heuristics tailored to the abstract domains we use which allows a significant boost in precision on such networks<br>
As an example on the widely used and public benchmark ACAS-XU, PyRAT proves all the defined properties in 292s while state of the art tools such as nnenum takes 319s and neurify 1265s.
</p>
</div>
......
......@@ -8,7 +8,7 @@
<div class="copyright">
<span>Copyright © 2021-{{ 'now' | date: "%Y" }} PyRAT. All Rights Reserved.</span>
</div>
<a href="{{ '/contact.html' | relative_url }}">Contact us</a>
<a href="{{ '/contact.html' | relative_url }}">Contact us</a> - <a href="{{ '/acknowledgment.html' | relative_url }}">Acknowledgments</a>
</div>
</div>
</div>
......
---
date: 16/04/2020
title: Nothing particular
logo: assets/images/local-rob.png
layout: new
---
Test news
\ No newline at end of file
---
date: 25/12/2021
title: Christmas
logo: assets/images/visualising.png
layout: new
---
Merry Christmas!
\ No newline at end of file
---
date: 05/02/1996
date: 01/09/2022 16:00
title: Website launched!
logo: assets/images/pyrat-logo.png
layout: new
---
PyRAT website is not yet officially launched !
\ No newline at end of file
The PyRAT website is officially online! Regular updates and news will be put in this website to inform of the latest progress in the tool.
To learn more about the tool visit the main page or <a href="{{ '/contact.html' | relative_url }}">contact us</a>
\ No newline at end of file
---
layout: default
title: PyRAT — Acknowledgments
---
<section class="section contact-page">
<div class="container">
<h1 class="title">Acknowledgments</h1>
<h3 style="margin-bottom: 1rem"> This work was supported by:</h3>
<div class="columns">
<div class="column is-one-third has-text-centered">
<a href="sparta.eu"><img width=100 src="{{'/assets/images/sparta-logo.png' | relative_url}}"></a>
</div>
<div class="column ack">
<p>
the SAFAIR subproject of the project SPARTA which has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 830892.
</p>
</div>
</div>
<div class="columns">
<div class="column is-one-third has-text-centered">
<a href="cps4eu.eu"><img width=100 src="{{'/assets/images/cps4eu-logo.png' | relative_url}}"></a>
</div>
<div class="column ack">
<p>
the CPS4EU project that has received funding from the ECSEL Joint Undertaking (JU) under grant agreement No 826276.
</p>
</div>
</div>
<div class="columns">
<div class="column is-one-third has-text-centered">
<a href="confiance.ai"><img width=100 src="{{'/assets/images/confiance-logo.jpg' | relative_url}}"></a>
</div>
<div class="column ack">
<p>
the confiance.ai research program, a French consortium with industrial and academic partners that aims to design and industrialize trustworthy AI-based critical systems.
</p>
</div>
</div>
<div class="columns ">
<div class="column is-one-third">
</div>
<div class="column ack">
<p>
the French Ministère des armées - Agence de l’innovation de défense (Ministry of Armed Forces - Defence Innovation Agency).
</p>
</div>
</div>
</div>
</section>
......@@ -10,7 +10,7 @@
}
.mainContent {
margin-top: 52px;
margin-top: 55px;
padding-bottom: 12rem;
}
......@@ -32,6 +32,11 @@ h2 {
margin-bottom: 2.75rem;
}
body {
position: relative;
min-height: 95vh;
}
/* NEWS
------------------------------------------------*/
......@@ -58,6 +63,16 @@ h2 {
/* FOOTER
------------------------------------------------------------------*/
.footer {
clear: both;
padding-top: 3rem;
padding-bottom: 1rem;
position: absolute;
bottom: 0;
width: 100%;
height: 12rem;
}
.copyright {
margin: auto;
margin-top: 1.35em;
......@@ -123,7 +138,8 @@ h2 {
padding-top: 2em;
}
/* Body section */
/* BODY SECTION
------------------------------------------------------------------ */
.verify .card {
border-top-left-radius: 1rem;
......@@ -152,3 +168,11 @@ h2 {
.key img {
max-width: 350px;
}
/* ACK Page
------------------------------------------------------------------ */
.column.ack {
display: flex;
align-items: center;
}
assets/images/confiance-logo.jpg

17.2 KiB

assets/images/cps4eu-logo.png

74.5 KiB

assets/images/local-rob.png

34.5 KiB | W: | H:

assets/images/local-rob.png

34.7 KiB | W: | H:

assets/images/local-rob.png
assets/images/local-rob.png
assets/images/local-rob.png
assets/images/local-rob.png
  • 2-up
  • Swipe
  • Onion skin
assets/images/sparta-logo.png

93.9 KiB

0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment