Lecture 21: Adversarial Networks CS109B Data Science 2 Pavlos - PowerPoint PPT Presentation

Lecture 21: Adversarial Networks CS109B Data Science 2 Pavlos Protopapas and Mark Glickman 1

How vulnerable are Neural Networks? Uses of Neural Networks CS109B, P ROTOPAPAS , G LICKMAN

How vulnerable are Neural Networks? CS109B, P ROTOPAPAS , G LICKMAN

Explaining Adversarial Examples [Goodfellow et. al ‘15] 1. Robust attacks with FGSM 2. Robust defense with Adversarial Training CS109B, P ROTOPAPAS , G LICKMAN

Explaining Adversarial Examples CS109B, P ROTOPAPAS , G LICKMAN

Some of these adversarial examples can even fool humans: CS109B, P ROTOPAPAS , G LICKMAN

Attacking with F ast G radient S ign M ethod (FGSM) W X L x + λ · sign( r x L) ) x ∗ CS109B, P ROTOPAPAS , G LICKMAN

x + λ · sign( r x L) ) x ∗ CS109B, P ROTOPAPAS , G LICKMAN

Defending with Adversarial Training 1. Generate adversarial examples 2. Adjust labels CS109B, P ROTOPAPAS , G LICKMAN

Defending with Adversarial Training “ Panda ” 1. Generate adversarial examples 2. Adjust labels CS109B, P ROTOPAPAS , G LICKMAN

Defending with Adversarial Training “ Panda ” 1. Generate adversarial examples 2. Adjust labels 3. Add them to the training set 4. Train new network CS109B, P ROTOPAPAS , G LICKMAN

Attack methods post GoodFellow 2015 ● FGSM [Goodfellow et. al ‘15] ● JSMA [Papernot et. al ‘16] ● C&W [Carlini + Wagner ‘16] ● Step-LL [Kurakin et. al ‘17] ● I-FGSM [Tramer et. al ‘18] CS109B, P ROTOPAPAS , G LICKMAN

White box attacks W L x + λ · sign( r x L) ) x ∗ x + λ · r x L ) x ∗ CS109B, P ROTOPAPAS , G LICKMAN

“ Black Box ” Attacks “ Black Box ” Attacks [Papernot et. al ‘17] CS109B, P ROTOPAPAS , G LICKMAN

“ Black Box ” Attacks Examine inputs and outputs of the model CS109B, P ROTOPAPAS , G LICKMAN

“ Black Box ” Attacks Panda CS109B, P ROTOPAPAS , G LICKMAN

“ Black Box ” Attacks Panda Gibbon CS109B, P ROTOPAPAS , G LICKMAN

“ Black Box ” Attacks Panda Gibbon Ostrich CS109B, P ROTOPAPAS , G LICKMAN

“ Black Box ” Attacks Train a model that performs the same as the black box CS109B, P ROTOPAPAS , G LICKMAN

“ Black Box ” Attacks Train a model that performs the same as the black box Panda Gibbon Ostrich CS109B, P ROTOPAPAS , G LICKMAN

“ Black Box ” Attacks Now attack the model you just trained with “ white ” box attack W L x + λ · r x L ) x ∗ x + λ · sign( r x L) ) x ∗ CS109B, P ROTOPAPAS , G LICKMAN

“ Black Box ” Attacks Use those adversarial examples to the “ black ” box CS109B, P ROTOPAPAS , G LICKMAN

CleverHans A Python library to benchmark machine learning systems' vulnerability to adversarial examples. https://github.com/tensorflow/cleverhans http://www.cleverhans.io/ CS109B, P ROTOPAPAS , G LICKMAN

More Defenses Mixup: Smooth decision boundaries: • Mix two training examples • Regularize the derivatives wrt to x Augment training set • x = λ x i + (1 − λ ) x j ˜ y = λ y i + (1 − λ ) y j ˜ P AVLOS P ROTOPAPAS

Physical attacks Object Detection • • Adversarial Stickers CS109B, P ROTOPAPAS , G LICKMAN

Thank you. CS109B, P ROTOPAPAS , G LICKMAN

Lecture 21: Adversarial Networks CS109B Data Science 2 Pavlos - PowerPoint PPT Presentation

Lecture 21: Adversarial Networks CS109B Data Science 2 Pavlos Protopapas and Mark Glickman 1 How vulnerable are Neural Networks? Uses of Neural Networks CS109B, P ROTOPAPAS , G LICKMAN How vulnerable are Neural Networks? CS109B, P ROTOPAPAS ,

CSC321 Lecture 19: Generative Adversarial Networks Roger Grosse Roger Grosse CSC321 Lecture 19:

Adversarial Training Attacks on Deep Networks and Generative Adversarial Networks Erkut Erdem

Machine Learning Lecture 13: Generative Adversarial Networks (I) Nevin L. Zhang

Adversarial Examples and Adversarial Training Ian Goodfellow, OpenAI Research Scientist Guest

Generative Adversarial Networks, Wasserstein Distance, and Adversarial Loss Zhiyu Min Alibaba

Advanced Machine Learning CS 7140 - Spring 2018 Lecture 20: Generative Adversarial Networks

Advanced Machine Learning CS 7140 - Spring 2019 Lecture 20: Generative Adversarial Networks

CS7015 (Deep Learning) : Lecture 23 Generative Adversarial Networks (GANs) Mitesh M. Khapra

CS7015 (Deep Learning) : Lecture 23 Generative Adversarial Networks (GANs) Mitesh M. Khapra

Generative Adversarial Networks Stefano Ermon, Aditya Grover Stanford University Lecture 10

Neglected topics CS 446 Adversarial examples and deep networks 1 / 23 Adversarial

CSC321 Lecture 22: Adversarial Learning Roger Grosse Roger Grosse CSC321 Lecture 22: Adversarial

CSC421/2516 Lecture 18: Generative Adversarial Networks Roger Grosse and Jimmy Ba Roger Grosse

LAB MEETING: A Connection Between Generative Adversarial Networks, Inverse Reinforcement Learning

Generative Adversarial Networks Stefano Ermon, Aditya Grover Stanford University Lecture 9

Generative Adversarial Networks Stefano Ermon, Aditya Grover Stanford University Lecture 9

Resources for Endangered Languages Specifications for a Roadmap Dafydd Gibbon U Bielefeld,

PROPERTY OWNERS MEETING SEWER SERVICE CHARGE RATES Burlingame Hills Sewer Maintenance District

Arizona Department of Environmental Quality Water Quality Improvement Grant Program Who am I?

Anchorage David Saalfeld Regional Wildlife Biologist Alaska Department of Fish and Game General

MENTAL WELLBEING: THE HEART OF YOUR TOTAL REWARD PROPOSITION Jane Gibbon Group Reward Director

City of Northfield Community S olar Garden S ubscription Approval August 8, 2017 Background

2016 Annual General Meeting of Shareholders Important notice and disclaimer Visit

The Streets of Adelaide BORN OF REFORM 1832 1836 By Dr Jeff Nicholas A comprehensive