Cap6412 21spring Towards Deep Learning

CAP6412 21Spring-Towards deep learning models resistant to adversarial attacks

So um today we're gonna be uh presenting this paper um uh uh

CAP6412 21Spring-Introduction Lecture -1

... of vertex we will learn so it's white boxer tech so there we so we are going to always take a class you know

CAP6412 21Spring-Explainable AI: Interpreting, Explaining and Visualizing Deep Learning

...

CAP6412 21Spring-Intriguing properties of neural networks

... of its conclusions uh back to core understandings of

CAP6412 21Spring-Fast is better than free: Revisiting adversarial training

... and ziko coulter and this paper was presented in the international conference on

CAP6412 21Spring- On adaptive attacks to adversarial example defenses

... of the um this way that that this way adversarial examples are more optimized

CAP6412 21Spring-Towards Evaluating the Robustness of Neural Networks

lecture 21 CAP6412

... ResNet architecture was to the stem Stem is nothing but the first layer of a

CAP6412 21Spring-Explaining and harnessing adversarial examples

So uh for the adversarial training of

CAP6412 21Spring-Contrastive Learning with Adversarial Examples

... so in machining we could uh think of

Tutorial 5: Explainability in Deep Learning

... of all

CAP6412 21Spring-Introduction Lecture -2

So last lecture we talked about brittleness of

CAP6412 21Spring-Adversarial Self-Supervised Contrastive Learning

Paper is uh adversarial self-supervised contrastive

CAP6412 21Spring-Feature denoising for improving adversarial robustness

... can lead to the network

CAP6412 21Spring-Robust Pre-Training by Adversarial Contrastive Learning

... robust

CAP6412 21Spring-Universal adversarial perturbations

... single small image perturbation that pools a state-of-the-art

CAP6412 21Spring jan25

Today i'll talk about

CAP6412 21Spring-Smoothgrad: removing noise by adding noise

... published in the international conference for