Really interesting paper exploring adversarial inputs to ML models:

They conclude:
* It's a property of the input data, not the training
* You can even train a model on non-robust features and obtain a model that works well on the original input data!

@wilfredh I'm not sure I'm drawing the same conclusion from their own examples...

@cwebber @wilfredh
I too overviewed that
Unfortunately ML is my weakest point,so I couldn't understand it deeply

Sign in to participate in the conversation

The original server operated by the Mastodon gGmbH non-profit