secml

secml: Secure and Explainable Machine Learning in Python

Code:

[secml @secml_py]

secml v0.14 is out!
https://t.co/j6MkKuk2IT
https://t.co/CnEDnkOjzj
https://t.co/UAizk3bEOb

Highlights:
- Foolbox attacks and RobustBench models now included, with notebooks
- New notebook tutorial with an application on Android Malware Detection

[Luca Demetrio @zangobot]

Update SecML Malware 0.2.6: support the SOREL DNN!
Thank you @rharang again for releasing it!
https://t.co/ZU6ANFPQtP

#secml #advml #malware #security #research

[secml @secml_py]

Now part of FoolBox too!

[Maura Pintor @maurapintor]

📢Very happy to announce that our paper "Fast Minimum-norm Adversarial Attacks through Adaptive Norm Constraints" was accepted for a poster presentation at @NeurIPSConf! This is joint work with @wielandbr, @biggiobattista, and Fabio Roli. 1/n

[Machine Learning Security Laboratory @mlsec_lab]

Join us for our event on Machine Learning Security! Tuesday, March 8th, 2022, at 16:00 CET.
Invited talk by Francesco Croce (University of Tübingen).

Registration: https://t.co/1Uhmnj2l0k

YT Live: https://t.co/Wex7Hd8WdB

#adversarial #machinelearning #ai #security #mlsec

[Machine Learning Security Laboratory @mlsec_lab]

Join us for our first seminar event on Machine Learning Security Tomorrow, Dec 7th, 2021, at 15.00 CET 🥳

Invited talk by David Stutz (Max Planck Institute for Informatics).

Registration here (free): https://t.co/Q947PiiLya

YT Live Stream: https://t.co/MGFLNyMPYj

[Machine Learning Security Laboratory @mlsec_lab]

We are excited to present our seminar series on Adversarial Machine Learning!

We will host David Stutz (Max Planck Institute) for our first event on Dec 7th at 15:00 CET.

Free registration here:
https://t.co/4VdV5e6yCX

#adversarial #machinelearning #ai #security

[secml @secml_py]

SecML + docker + GUI... it's PandaVision! Cool work by @maurapintor

[Maura Pintor @maurapintor]

Dream: evaluating adversarial robustness (with attack configuration!) without writing code

Reality: https://t.co/pIEbhJFbHI

Check out my recent work, 🌟 the repo if you like it!

Preview: Integrate and automate security evaluations with ONNX, PyTorch, and SecML!

Video in 🧵!

[Ambra @ambrademontis]

The second lecture of the course "Machine Learning Security" is on YouTube: https://t.co/dVvRtpRbde

Topic: Adversarial Examples and defenses.

Lecturer: @biggiobattista, @zangobot

#MLSec #MachineLearning #AI #adversarial #Malware @adversarial_ML @trustworthy_ml @aivillage_dc

[secml @secml_py]

New README is UP: https://t.co/CnEDnkOjzj!

[Battista Biggio @biggiobattista]

We're preparing a short course for PhD students on machine learning security, and open sourcing the content. Any feedback is more than welcome -- towards improving next year's extended edition!

https://t.co/aXbkRzaR0V

[secml @secml_py]

Refactoring is ongoing. Fasten your seatbelts as we'll have fun soon! And star our github repo if you like secml!

https://t.co/CnEDnkOjzj

[secml @secml_py]

@icymi_py @pluribus_one thanks for sharing!

[Pluribus One @pluribus_one]

Our article for the @AssureMOSSH2020 WP3, dedicated to continuos analysis and correction of secure code

@secml_py (Secure ML Library) is an #opensource #Python library for the security evaluation of #MachineLearning algorithms

#cybersecurity #ai #ml @MarcoMelisIT @ambrademontis

[Maura Pintor @maurapintor]

ALOHA defines a framework for optimizing the design of Deep Learning systems on heterogeneous low-energy computing platforms, and includes adversarial robustness evaluation with @secml_py
. Check out the workshop where we show the achievements of this project!

[secml @secml_py]

@pierrep67 Thanks a lot! :) Major updates are coming after the summer! ;)

[secml @secml_py]

@ram_ssk Thanks Ram! :)

[Maura Pintor @maurapintor]

Find efficiently minimum-norm Adversarial Examples in different L-p norms with FMN! 🙌

Updated (increasing) list of available implementations: https://t.co/VLEYLRilQH

@biggiobattista @wielandbr

[Battista Biggio @biggiobattista]

This is amazing. @FSecure explains poisoning attacks using @secml_py ! Thanks! :)

[secml @secml_py]

Please do give us a star if you like SecML! https://t.co/CnEDnkOjzj

[secml @secml_py]

Check out this new attack. You'll be surprised ;)

[Battista Biggio @biggiobattista]

📍Paper updated:

Fast Minimum-norm Adversarial Attacks through Adaptive Norm Constraints

Paper: https://t.co/pexKieIgSS
Github: https://t.co/ymUa0iwGYT

Available within @secml_py and Foolbox. w/ @maurapintor @wielandbr

[secml @secml_py]

Our model zoo is growing!

[Marco Melis @MarcoMelisIT]

This week we added new #pretrained #models for @secml in our zoo:
- Support Vector Machines and our Secure SVM from https://t.co/QhsXBZGGvy, on the Drebin #android ds
- #pytorch AlexNet + SVMs from https://t.co/ogQyXLJFgQ, on the #iCub ds

Check them out: https://t.co/7hXBYcFNim