Fwd: Member requesting to join cleverhans dev

[Ian Goodfellow]

Answering the message sent by neurocat.fg upon membership request:

Pull requests on all these topics are welcome.

So far we try to mostly offer code related to making attacks or defenses without committing to providing a full model architecture and training stack. In the future we may need to start including more of the architecture and training stack in order to provide useful implementations of the defenses.

---------- Forwarded message ----------
From: cleverhans dev <cleverhans-...@googlegroups.com>
Date: Mon, Jun 18, 2018 at 7:27 AM
Subject: Member requesting to join cleverhans dev
To: goodf...@google.com

cleverhans dev

Google Groups

(unknown) (neuro...@gmail.com) has asked to join the cleverhans dev group.

Message from (unknown)

Hey Guys, congratulations to your new release. Is there any capacity planed on extending the architecture of cleverhans for modules like criterias or different math. metrics for the enlarging the scope of an attack in further versions e.g. https://github.com/bethgelab/foolbox. Is there any work capacity planed to implement different input dimensions like (audio, text) or output dimensions/lossfunctions? Can you put the scope of cleverhans in short words? Is it just an environment for you guys implementing your research findings in a modular way. I guess in the future different metrics will play a huge role for provable adversary. Is there space for architectural contributions? Kind regards Florens

Manage members

Start a new group. Visit the help center.

[neuro...@gmail.com]

Thanks for the response Ian,

I'll stay tuned and give as much as my capacity allows.