DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY - PowerPoint PPT Presentation

DATA ANALYTICS USING DEEP LEARNING GT 8803 // FALL 2019 // JOY ARULRAJ L E C T U R E # 2 0 : A D V E R S A R I A L T R A I N I N G

administrivia • Reminders – Best project prize – Quiz cancelled – Guest lecture GT 8803 // Fall 2019 2

CREDITS • Slides based on a lecture by: – Ian Goodfellow @ Google Brain GT 8803 // Fall 2019 3

OVERVIEW • What are adversarial examples? • Why do they happen? • How can they be used to compromise machine learning systems? • What are the defenses? • How to use adversarial examples to improve machine learning (even without adversary)? GT 8803 // Fall 2019 4

ADVERSARIAL EXAMPLES 5 GT 8803 // Fall 2019

Since 2013, deep neural networks have matched human performance at... ...recognizing objects and faces…. (Szegedy et al, 2014) (Taigmen et al, 2013) ...solving CAPTCHAS and reading addresses... (Goodfellow et al, 2013) (Goodfellow et al, 2013) and other tasks... 6 GT 8803 // Fall 2019

Adversarial Examples Timeline: “Adversarial Classification” Dalvi et al 2004: fool spam filter “Evasion Attacks Against Machine Learning at Test Time” Biggio 2013: fool neural nets Szegedy et al 2013: fool ImageNet classifiers imperceptibly Goodfellow et al 2014: cheap, closed form attack 7 GT 8803 // Fall 2019

Turning Objects into “Airplanes” 8 GT 8803 // Fall 2019

Attacking a Linear Model 9 GT 8803 // Fall 2019

Not just for neural nets • Linear models – Logistic regression – Softmax regression – SVMs • Decision trees • Nearest neighbors GT 8803 // Fall 2019 10

Adversarial Examples from Overfitting O O x x O O x x GT 8803 // Fall 2019 11

Adversarial Examples from Overfitting O O O O x x x O x GT 8803 // Fall 2019 12

Modern deep nets are very piecewise linear Rectified linear unit Maxout Carefully tuned sigmoid LSTM GT 8803 // Fall 2019 13

Nearly Linear Responses in Practice GT 8803 // Fall 2019 14

Small inter-class distances Corrupted Clean Perturbation example example Perturbation changes the true class Random perturbation does not change the class Perturbation changes the input to “rubbish class” All three perturbations have L2 norm 3.96 This is actually small. We typically use 7! GT 8803 // Fall 2019 15

The Fast Gradient Sign Method GT 8803 // Fall 2019 16

Maps of Adversarial and Random Cross-Sections (collaboration with David Warde-Farley and Nicolas Papernot) GT 8803 // Fall 2019 17

Maps of Adversarial Cross-Sections GT 8803 // Fall 2019 18

Maps of RANDOM Cross-Sections (collaboration with David Warde-Farley and Nicolas Papernot) GT 8803 // Fall 2019 19

Estimating the Subspace Dimensionality GT 8803 // Fall 2019 20

Clever Hans (“Clever Hans, Clever Algorithms,” Bob Sturm) GT 8803 // Fall 2019 21

Wrong almost everywhere GT 8803 // Fall 2019 22

Adversarial Examples for RL (Huang et al., 2017) GT 8803 // Fall 2019 23

High-Dimensional Linear Models Clean examples Adversarial examples Weights Signs of weights GT 8803 // Fall 2019 24

Linear Models of ImageNet (Andrej Karpathy, “Breaking Linear Classifiers on ImageNet”) GT 8803 // Fall 2019 25

RBFs behave more intuitively GT 8803 // Fall 2019 26

Cross-model, cross-dataset generalization GT 8803 // Fall 2019 27

Cross-technique transferability (Papernot 2016) GT 8803 // Fall 2019 28

Transferability Attack Target model with Substitute model Train your own model unknown weights, mimicking target machine learning model with known, algorithm, training differentiable function set; maybe non-differentiable Deploy adversarial Adversarial crafting examples against the against substitute target; transferability property results in them Adversarial succeeding examples GT 8803 // Fall 2019 29

(Papernot 2016) 30 GT 8803 // Fall 2019

Enhancing Transfer With Ensembles (Liu et al, 2016) GT 8803 // Fall 2019 31

Adversarial Examples in the Human Brain These are concentric circles, not intertwined spirals. (Pinna and Gregory, 2002) GT 8803 // Fall 2019 32

Practical Attacks • Fool real classifiers trained by remotely hosted API (MetaMind, Amazon, Google) • Fool malware detector networks • Display adversarial examples in the physical world and fool machine learning systems that perceive them through a camera GT 8803 // Fall 2019 33

Adversarial Examples in the Physical World GT 8803 // Fall 2019 34

Failed defenses Generative Removing perturbation pretraining with an autoencoder Adding noise at test time Ensembles Confidence-reducing Error correcting perturbation at test time codes Multiple glimpses Weight decay Double backprop Adding noise Various at train time Dropout non-linear units GT 8803 // Fall 2019 35

Generative Modeling is not SufficienT GT 8803 // Fall 2019 36

Universal approximator theorem Neural nets can represent either function: Maximum likelihood doesn’t cause them to learn the right function. But we can fix that... GT 8803 // Fall 2019 37

ADVERSARIAL TRAINING 38 GT 8803 // Fall 2019

Training on Adversarial Examples GT 8803 // Fall 2019 39

Adversarial Training of other Models • Linear models: SVM / linear regression cannot learn a step function, so adversarial training is less useful, very similar to weight decay • k-NN: adversarial training is prone to overfitting. • Takeway: neural nets can actually become more secure than other models. Adversarially trained neural nets have the best empirical success rate on adversarial examples of any machine learning model. GT 8803 // Fall 2019 40

Weaknesses Persist GT 8803 // Fall 2019 41

Adversarial Training Labeled as bird Still has same label (bird) Decrease probability of bird class GT 8803 // Fall 2019 42

Virtual Adversarial Training Unlabeled; model New guess should guesses it’s probably match old guess a bird, maybe a plane (probably bird, maybe plane) Adversarial perturbation intended to change the guess GT 8803 // Fall 2019 43

Text Classification with VAT RCV1 Misclassification Rate 8.00 7.70 7.50 7.40 7.20 7.12 7.00 7.05 6.97 6.68 6.50 6.00 Earlier SOTA Our baseline Virtual Both + Adversarial bidirectional model Zoomed in for legibility GT 8803 // Fall 2019 44

Universal engineering machine (model-based optimization) Make new inventions by finding input that Training data Extrapolation maximizes model’s predicted performance 45 GT 8803 // Fall 2019

cleverhans Open-source library available at: https://github.com/openai/cleverhans Built on top of TensorFlow (Theano support anticipated) Standard implementation of attacks, for adversarial training and reproducible benchmarks 46 GT 8803 // Fall 2019

Conclusion • Attacking is easy • Defending is difficult • Adversarial training provides regularization and semi-supervised learning • The out-of-domain input problem is a bottleneck for model-based optimization generally GT 8803 // Fall 2019 47