Trusting Machine Learning Algorithms for Safeguards Applications

Slides:



Advertisements
Similar presentations
Supervised and Unsupervised learning and application to Neuroscience Cours CA6b-4.
Advertisements

Robust Real-time Object Detection by Paul Viola and Michael Jones ICCV 2001 Workshop on Statistical and Computation Theories of Vision Presentation by.
Attention Deficit Hyperactivity Disorder (ADHD) Student Classification Using Genetic Algorithm and Artificial Neural Network S. Yenaeng 1, S. Saelee 2.
Convolutional Neural Networks for Image Processing with Applications in Mobile Robotics By, Sruthi Moola.
Hurieh Khalajzadeh Mohammad Mansouri Mohammad Teshnehlab
An Example of Course Project Face Identification.
Analysis of Classification Algorithms In Handwritten Digit Recognition Logan Helms Jon Daniele.
Detecting New a Priori Probabilities of Data Using Supervised Learning Karpov Nikolay Associate professor NRU Higher School of Economics.
Machine Learning Artificial Neural Networks MPλ ∀ Stergiou Theodoros 1.
Introduction to Machine Learning, its potential usage in network area,
Course Outline (6 Weeks) for Professor K.H Wong
Classify A to Z Problem Statement Technical Approach Results Dataset
Software Defects Cmpe 550 Fall 2005
2/13/2018 4:38 AM © Microsoft Corporation. All rights reserved. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN.
Big data classification using neural network
Machine Learning for Computer Security
Machine Learning – Classification David Fenyő
Inter-experimental LHC Machine Learning Working Group Activities
Applying Deep Neural Network to Enhance EMPI Searching
Deep Learning Amin Sobhani.
an introduction to: Deep Learning
Data Mining, Neural Network and Genetic Programming
Data Mining, Neural Network and Genetic Programming
From Vision to Grasping: Adapting Visual Networks
Saliency-guided Video Classification via Adaptively weighted learning
Part-Based Room Categorization for Household Service Robots
Evaluating a Real-time Anomaly-based IDS
Neural networks (3) Regularization Autoencoder
Classification of Hand-Written Digits Using Scattering Convolutional Network Dongmian Zou Advisor: Professor Radu Balan.
Poisoning Attacks with Back-Gradient Optimization
Azure Machine Learning Noam Brezis Madeira Data Solutions
Deep learning and applications to Natural language processing
AI in Cyber-security: Examples of Algorithms & Techniques
FaceNet A Unified Embedding for Face Recognition and Clustering
R-CNN region By Ilia Iofedov 11/11/2018 BGU, DNN course 2016.
Dynamic Routing Using Inter Capsule Routing Protocol Between Capsules
Advanced Analytics. Advanced Analytics What is Machine Learning?
Zan Gao, Deyu Wang, Xiangnan He, Hua Zhang
Bird-species Recognition Using Convolutional Neural Network
Feature Engineering Studio Special Session
Toward improved document classification and retrieval
Reflections on GDPR Article 15
Introduction to Deep Learning with Keras
INF 5860 Machine learning for image classification
Alain Goossens & Jean-Pierre Van Loo Data scientists – SII Belgium
Object Detection + Deep Learning
Pose Estimation for non-cooperative Spacecraft Rendevous using CNN
Object Classes Most recent work is at the object level We perceive the world in terms of objects, belonging to different classes. What are the differences.
Machine Learning 101 Intro to AI, ML, Deep Learning
Creating Data Representations
Stealing DNN models: Attacks and Defenses
Lecture: Deep Convolutional Neural Networks
Outline Background Motivation Proposed Model Experimental Results
Tuning CNN: Tips & Tricks
Explainable Machine Learning
Earthen Mounds Recognition Using LiDAR Images
Neural networks (3) Regularization Autoencoder
Bioinformatics 김유환, 문현구, 정태진, 정승우.
Deep Learning Authors: Yann LeCun, Yoshua Bengio, Geoffrey Hinton
An introduction to: Deep Learning aka or related to Deep Neural Networks Deep Structural Learning Deep Belief Networks etc,
Department of Computer Science Ben-Gurion University of the Negev
DRC with Deep Networks Tanmay Lagare, Arpit Jain, Luis Francisco,
Learning and Memorization
Machine Learning.
Peng Cui Tsinghua University
Report 2 Brandon Silva.
Machine Learning.
Jianbo Chen*, Le Song†✦, Martin J. Wainwright*◇ , Michael I. Jordan*
An introduction to neural network and machine learning
Presented By: Firas Gerges (fg92)
Presentation transcript:

Trusting Machine Learning Algorithms for Safeguards Applications Nathan Shoman SAND2019-2539 C

Many different commercial machine learning applications Images/logos copyright to their respective owners

Machine learning can be applied to domains relevant to safeguards Anomaly detection in multivariate data sets Zhang (2018) https://arxiv.org/abs/1811.08055 Anomaly detection in images Neural Network Prediction Actual frame Anomaly UCSD Dataset

Understanding neural networks Explain forward-pass, back-prop, classification, supervision http://cs231n.github.io/neural-networks-1/ Stanford CS231n (2018) http://cs231n.github.io/

More complex networks become difficult to interpret Inception network Szegedy et al (2015) https://arxiv.org/abs/1409.4842

Practical considerations for evaluating NN performance Precision / recall / F1 Importance of validation and test data Exploring intermediate layers Layer 5 Layer 3 https://arxiv.org/abs/1311.2901 http://cs231n.github.io/understanding-cnn/ https://arxiv.org/pdf/1312.6034.pdf Zeiler (2013) https://arxiv.org/abs/1311.2901

Using LIME (Local Interpretable Model-agnostic Explanations) General algorithm to explain predictions of classifiers or regressors by approximating it locally with an interpretable model Fidelity – Interpretability Trade-off Riberio, et al. (2016) https://arxiv.org/abs/1602.04938

Using LIME with CNNs for image recognition and classification Riberio, et al. (2016) https://arxiv.org/abs/1602.04938

One Pixel Attack for Fooling Deep Neural Networks Su et al. (2017) https://arxiv.org/abs/1710.08864

Conclusions ML algorithms are powerful tools that could improve existing safeguards and security systems Trust of machine learning algorithms is essential to acceptance by the safeguards community Analysis with tools such as LIME is important when presenting results Newly developed strategies such as Layer-wise Relevance Propagation (Binder, et al. 2016) and Testing with Concept Activation Vectors (Kim et al. 2018) can provide further insight into ML classification logic One pixel attack is detectable, even when not perceived via human eye, but require extra pre-processing (Xu et al. 2017, Liang et al. 2017) Binder – 1604.00825 Kim - http://proceedings.mlr.press/v80/kim18d/kim18d.pdf Xu-1704.01155 Liang-1705.08378