Theoretical Analysis of Adversarial Learning: A Minimax Approach - PowerPoint PPT Presentation

Theoretical Analysis of Adversarial Learning: A Minimax Approach Zhuozhuo Tu 1 , Jingwei Zhang 2,1 , Dacheng Tao 1 1 The University of Sydney 2 The Hong Kong University of Science and Technology NeurIPS 2019

� � � � � � � � � � � � � � � � � �� x ′ ∈ N ( x ) l ( h ( x ′ ) , y )] E ( x,y ) ∼ P [ �� ◮ �� l q �� ◮ ��

� � � � � � � � � � � � � � � � ��

� � � � � � � � � � � � � � � � �� R P ( h ) = E ( x,y ) ∼ P [ l ( h ( x ) , y )] .

� � � � � � � � � � � � � � � � � �� R P ( h ) = E ( x,y ) ∼ P [ l ( h ( x ) , y )] . x ′ ∈ N ( x ) l ( h ( x ′ ) , y )] , R P ( h, B ) = E ( x,y ) ∼ P [ �� N ( x ) = { x ′ : x ′ − x ∈ B} �

� � � � � � � � � � � � � � � ��

� � � � � � � � � � � � � � � � �� T h : Z → Z �� R P ( h, B ) = R P ′ ( h )

� � � � � � � � � � � � � � � � � �� T h : Z → Z �� R P ( h, B ) = R P ′ ( h ) �� P ′ �� P W p ( P, P ′ ) ≤ � B

� � � � � � � � � � � � � � � � � �� T h : Z → Z �� R P ( h, B ) = R P ′ ( h ) �� P ′ �� P W p ( P, P ′ ) ≤ � B R P ( h, B ) ≤ R � B , 1 ( P, h ) , ∀ h ∈ H

� � � � � � � � � � � � � � � � ��

� � � � � � � � � � � � � � � � � � �� f ∈ F �� z ∈ Z � �� λ f,z �� f ( z ′ ) − f ( z ) ≤ λ f,z d Z ( z, z ′ ) �� z ′ ∈ Z �

Theoretical Analysis of Adversarial Learning: A Minimax Approach - PowerPoint PPT Presentation

Theoretical Analysis of Adversarial Learning: A Minimax Approach Zhuozhuo Tu 1 , Jingwei Zhang 2,1 , Dacheng Tao 1 1 The University of Sydney 2 The Hong Kong University of Science and Technology NeurIPS 2019

Adversarial Games Lectures Contents 1. Introduction 2. Move Evaluation using MiniMax

Outline Games Perfect play: principles of adversarial search minimax decisions

CSE 473: Artificial Intelligence Today Spring 2012 Adversarial Search Minimax search

SECURITY, ADVERSARIAL SECURITY, ADVERSARIAL LEARNING, AND PRIVACY LEARNING, AND PRIVACY

Outline Games Perfect play minimax decisions Adversarial Search pruning

CS3750: ADVANCED MACHINE LEARNING GENERATIVE ADVERSARIAL NETWORKS Adapted from Slides made by

Adversarial Search Problem: minimax takes too long. Solution: improve algorithm to ignore

CS885 Reinforcement Learning Lecture 13c: June 13, 2018 Adversarial Search [RusNor] Sec. 5.1-5.4

Adversarial Learning Bounds for Linear Classes and Neural Nets Understanding Adversarial Learning

Deep Adversarial Learning for NLP 9:00 10:30 Introduction and Adversarial Training, GANs

Minimax Statistical Learning with Wasserstein distances Jaeho Lee & Maxim Raginsky NeurIPS

A THEORETICAL LOOKS AT ADVERSARIAL EXAMPLES Tom Goldstein and also Ali Shafahi, Ronny

CSE 573: Artificial Intelligence Adversarial Search Dan Weld Based on slides from Dan Klein,

Adversarial Approaches to Bayesian Learning and Bayesian Approaches to Adversarial Robustness

Friendly Adversarial Training: Attacks Which Do Not Kill Training Make Adversarial Learning

Minimax-Angle Learning for Optimal Treatment Decision with Heterogeneous Data Chengchun Shi

How Fast Can Higher-Order Masking Be in Software? Dahmun Goudarzi and Matthieu Rivain EUROCRYPT

De-biasing arbitrary convex regularizers and asymptotic normality Pierre C Bellec, Rutgers

IEEE IAS Technical Books Coordinating Committee (IAS/TBCC) 2013 I&CPS Lisa Perry IEEE-SA

New matrix norms for structured matrix estimation Jean-Philippe Vert Optimization and Statistical

IPv6 Multicast Over TEIN Pujan Srivastava | pujan@ait.asia Asian Institute of Technology

Cool gas inside early-type galaxies galaxies Timothy A.

Asteroid color photometry with Gaia and synergies with other space missions Marco Delbo

UMBC A B M A L T F O U M B C I M Y O R T 1 (12/11/06) I E S R C E O V

Theoretical Analysis of Adversarial Learning: A Minimax Approach - PowerPoint PPT Presentation

Theoretical Analysis of Adversarial Learning: A Minimax Approach Zhuozhuo Tu 1 , Jingwei Zhang 2,1 , Dacheng Tao 1 1 The University of Sydney 2 The Hong Kong University of Science and Technology NeurIPS 2019

Adversarial Games Lectures Contents 1. Introduction 2. Move Evaluation using MiniMax

Outline Games Perfect play: principles of adversarial search minimax decisions

CSE 473: Artificial Intelligence Today Spring 2012 Adversarial Search Minimax search

SECURITY, ADVERSARIAL SECURITY, ADVERSARIAL LEARNING, AND PRIVACY LEARNING, AND PRIVACY

Outline Games Perfect play minimax decisions Adversarial Search pruning

CS3750: ADVANCED MACHINE LEARNING GENERATIVE ADVERSARIAL NETWORKS Adapted from Slides made by

Adversarial Search Problem: minimax takes too long. Solution: improve algorithm to ignore

CS885 Reinforcement Learning Lecture 13c: June 13, 2018 Adversarial Search [RusNor] Sec. 5.1-5.4

Adversarial Learning Bounds for Linear Classes and Neural Nets Understanding Adversarial Learning

Deep Adversarial Learning for NLP 9:00 10:30 Introduction and Adversarial Training, GANs

Minimax Statistical Learning with Wasserstein distances Jaeho Lee &amp; Maxim Raginsky NeurIPS

A THEORETICAL LOOKS AT ADVERSARIAL EXAMPLES Tom Goldstein and also Ali Shafahi, Ronny

CSE 573: Artificial Intelligence Adversarial Search Dan Weld Based on slides from Dan Klein,

Adversarial Approaches to Bayesian Learning and Bayesian Approaches to Adversarial Robustness

Friendly Adversarial Training: Attacks Which Do Not Kill Training Make Adversarial Learning

Minimax-Angle Learning for Optimal Treatment Decision with Heterogeneous Data Chengchun Shi

How Fast Can Higher-Order Masking Be in Software? Dahmun Goudarzi and Matthieu Rivain EUROCRYPT

De-biasing arbitrary convex regularizers and asymptotic normality Pierre C Bellec, Rutgers

IEEE IAS Technical Books Coordinating Committee (IAS/TBCC) 2013 I&amp;CPS Lisa Perry IEEE-SA

New matrix norms for structured matrix estimation Jean-Philippe Vert Optimization and Statistical

IPv6 Multicast Over TEIN Pujan Srivastava | pujan@ait.asia Asian Institute of Technology

Cool gas inside early-type galaxies galaxies Timothy A.

Asteroid color photometry with Gaia and synergies with other space missions Marco Delbo

UMBC A B M A L T F O U M B C I M Y O R T 1 (12/11/06) I E S R C E O V

Minimax Statistical Learning with Wasserstein distances Jaeho Lee & Maxim Raginsky NeurIPS

IEEE IAS Technical Books Coordinating Committee (IAS/TBCC) 2013 I&CPS Lisa Perry IEEE-SA