SlideShare a Scribd company logo
1 of 53
Introduction to Interpretable
Machine Learning
Presented by Giang Nguyen
KAIST, Nov 2019
Terminologies
- Interpretable ML
- Explanable AI
- X – AI
2
BIRD VIEW OVER EXPLAINABLE AI
Deep Learning as Blackbox
While powerful, deep learning models are difficult to interpret, and thus
often treated as a blackbox.
4
Interpretability
Interpretation is the process of giving explanations to humans.
Interpretability is not a well-defined concept
5
Types of Interpretable Methods
We can interpret the model either before building the model, when
building it, or after building a model.
Most interpretation methods for DNNs interpret the model after it is built.
6
Interpretation When Building a Model
7
Using Inherently Interpretable Models
(Sparse) linear models and decision trees are inherently interpretable.
8
Attention Mechanisms
Attention mechanisms guide deep neural networks to focus on
relevant input features, which allows to interpret how the model made
certain predictions.
9
[Bahdanau et al. 15] Neural Machine Translation by Jointly Learning to Align and Translate, ICLR 2015
Limitation of Conventional Attention Mechanisms
Conventional attention models may allocate attention inaccurately since
they are trained in a weakly-supervised manner.
The problem becomes more prominent when a task has no one-to-one
mapping from inputs to the final predictions.
10
Limitation of Conventional Attention Mechanisms
This is because the conventional attention mechanisms do not consider
uncertainties in the model and the input, which often leads to
overconfident attention allocations.
Such unreliability may lead to incorrect predictions and/or interpretations
which can result in fatal consequences for safety-critical applications.
11
Uncertainty Aware Attention (UA)
12
Uncertainty Aware Attention (UA)
13
Multi-class classification performance on the three health records datasets
Info-GAN
14
There are structures in the noise vectors that have meaningful and
consistent effects on the output of the generator.
However, there’s no systematic way to find these structures. The only
thing affecting to the generator output is the noise input, so we have no
idea how to modify the noise to generate expected images.
Info-GAN
15
The idea is to provide a latent code, which has meaningful and consistent
effects on the output - disentangled representation
The hope is that if you keep the code the same and randomly change the
noise, you get variations of the same digit.
Info-GAN
16
c1 ∼ Cat(K = 10, p = 0.1)
Interpretation After Building a Model
17
Understanding Black-Box Predictions
Given a high-accuracy blackbox model and a prediction from it, can we
answer why the model made a certain prediction?
[Koh and Liang 17] tackles this question by training a model’s prediction through its learning algorithm
and back to the training data.
To formalize the impact of a training point on a prediction, they ask the counterfactual:
What would happen if we did not have this training point or if its value were slightly changed?
18
[Koh and Liang 17] Understanding Black-box Predictions via Influence Functions, ICML 2017
Interpretable Mimic Learning
This framework is mainly based on knowledge distillation from Neural
Networks.
However, they use Gradient Boosting Trees (GBT) instead of another neural
network as the student model since GBT satisfies our requirements for
both learning capacity and interpretability.
19[Che et al. 2016] Z. Che, S. Purushotham, R. Khemani, and Y. Liu. Interpretable Deep Models for
ICU outcome prediction, AMIA 2016.
Knowledge distillation
G. Hinton et al. 15
Interpretable Mimic Learning
The resulting simple model works even better than the best deep learning
model – perhaps due to suppression of the overfitting.
20[Che et al. 2016] Z. Che, S. Purushotham, R. Khemani, and Y. Liu. Interpretable Deep Models for
ICU outcome prediction, AMIA 2016.
Visualizing Convolutional Neural Networks
Propose Deconvolution Network (deconvnet) to inversely map the feature
activations to pixel space and provide a sensitivity analysis to point out
which regions of an image affect to decision making process the most.
21
[Zeiler and Fergus 14] Visualizing and Understanding Convolutional Networks, ECCV 2014
Prediction difference analysis
22
The visualization method shows which pixels of a specific input image are
evidence for or against a prediction
[Zintgraf et al. 2017] Visualizing Deep Neural Network Decisions: Prediction Difference Analysis, ICLR 2017
Shown is the evidence for (red) and against (blue) the prediction.
We see that the facial features of the cockatoo are most supportive for the decision, and
parts of the body seem to constitute evidence against it.
Interpretation Before Building a Model
23
Understanding Data Through Examples
[Kim et al. 16] propose to interpret the given data by providing examples
that can show the full picture – majorities + minorities
[Kim et al. 16] Examples are not Enough, Learn to Criticize! Criticism for Interpretability 24
INTRODUCTION
AI is data-driven, what we get is what we have
26
Understanding data through examples
27
Understanding data through examples
28
Understanding data through examples
29
Understanding data through examples
30
Understanding data through examples
31
Understanding data through examples
33
Understanding data through examples
34
Ideas of paper
35
Related Work
37
Over-generalization
Over-generalization is consistent with evolutionary
theory [Zebrowitz ‘10, Schaller’ 06]
algorithms can help against over-generalization
38
Venn diagram of related works
MMD-critic
40
Approach
41
Maximum Mean Discrepancy (MMD)
42
MMD-Critic: Learning Prototypes and Criticisms
Experiments
44
Results
45
Prototype-based classification
• Use the learned prototypes for classification (nearest-neighbor)
46
Example Prototypes and Criticisms
• USPS Digits Dataset
Unrecognizable
47
Example Prototypes and Criticisms
• ImageNet Dataset – 2 breeds of dog
48
Pilot study with human subjects
Definition of interpretability: A method is interpretable if a user can
correctly and efficiently predict the method’s results.
Task: Assign a new data point to one of the groups using 1) all images
2) prototypes 3) prototypes and criticisms 4) small set of randomly
selected images
49
Pilot study with human subjects
Conclusion
Take-home messages
51
• There are three types of Interpretable Methods, but mostly after building
models
• Criticism and prototypes are equally important and are a step towards
improving interpretability of complex data distributions
• MMD-critic learns prototypes + criticisms that highlight aspects of
data that are overlooked by prototypes.
Discussion
52
• If we have the insight into a dataset, can we really build a better model?
Human intuition is biased and not realiable!
Gap in Interpretable ML research
53
• Limited work to explain the operation of RNNs, only CNN. Attention
mechanism is not enough. Especially in multimodal network (CNN +
RNN), this kind of research is more necessary
Thank you for your attention!

More Related Content

What's hot

Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)Hayim Makabee
 
Explainable AI in Industry (KDD 2019 Tutorial)
Explainable AI in Industry (KDD 2019 Tutorial)Explainable AI in Industry (KDD 2019 Tutorial)
Explainable AI in Industry (KDD 2019 Tutorial)Krishnaram Kenthapadi
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine LearningSri Ambati
 
An Introduction to XAI! Towards Trusting Your ML Models!
An Introduction to XAI! Towards Trusting Your ML Models!An Introduction to XAI! Towards Trusting Your ML Models!
An Introduction to XAI! Towards Trusting Your ML Models!Mansour Saffar
 
Stable Diffusion path
Stable Diffusion pathStable Diffusion path
Stable Diffusion pathVitaly Bondar
 
ブラックボックスからXAI (説明可能なAI) へ - LIME (Local Interpretable Model-agnostic Explanat...
ブラックボックスからXAI (説明可能なAI) へ - LIME (Local Interpretable Model-agnostic Explanat...ブラックボックスからXAI (説明可能なAI) へ - LIME (Local Interpretable Model-agnostic Explanat...
ブラックボックスからXAI (説明可能なAI) へ - LIME (Local Interpretable Model-agnostic Explanat...西岡 賢一郎
 
Classification Based Machine Learning Algorithms
Classification Based Machine Learning AlgorithmsClassification Based Machine Learning Algorithms
Classification Based Machine Learning AlgorithmsMd. Main Uddin Rony
 
Interpretability beyond feature attribution quantitative testing with concept...
Interpretability beyond feature attribution quantitative testing with concept...Interpretability beyond feature attribution quantitative testing with concept...
Interpretability beyond feature attribution quantitative testing with concept...MLconf
 
Machine learning
Machine learningMachine learning
Machine learningeonx_32
 
Machine Learning Unit 1 Semester 3 MSc IT Part 2 Mumbai University
Machine Learning Unit 1 Semester 3  MSc IT Part 2 Mumbai UniversityMachine Learning Unit 1 Semester 3  MSc IT Part 2 Mumbai University
Machine Learning Unit 1 Semester 3 MSc IT Part 2 Mumbai UniversityMadhav Mishra
 
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)Krishnaram Kenthapadi
 
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...Sri Ambati
 
Intro to modelling-supervised learning
Intro to modelling-supervised learningIntro to modelling-supervised learning
Intro to modelling-supervised learningJustin Sebok
 
Introduction to ML (Machine Learning)
Introduction to ML (Machine Learning)Introduction to ML (Machine Learning)
Introduction to ML (Machine Learning)SwatiTripathi44
 
The fundamentals of Machine Learning
The fundamentals of Machine LearningThe fundamentals of Machine Learning
The fundamentals of Machine LearningHichem Felouat
 

What's hot (20)

Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)
 
Shap
ShapShap
Shap
 
Explainable AI in Industry (KDD 2019 Tutorial)
Explainable AI in Industry (KDD 2019 Tutorial)Explainable AI in Industry (KDD 2019 Tutorial)
Explainable AI in Industry (KDD 2019 Tutorial)
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learning
 
An Introduction to XAI! Towards Trusting Your ML Models!
An Introduction to XAI! Towards Trusting Your ML Models!An Introduction to XAI! Towards Trusting Your ML Models!
An Introduction to XAI! Towards Trusting Your ML Models!
 
Stable Diffusion path
Stable Diffusion pathStable Diffusion path
Stable Diffusion path
 
ブラックボックスからXAI (説明可能なAI) へ - LIME (Local Interpretable Model-agnostic Explanat...
ブラックボックスからXAI (説明可能なAI) へ - LIME (Local Interpretable Model-agnostic Explanat...ブラックボックスからXAI (説明可能なAI) へ - LIME (Local Interpretable Model-agnostic Explanat...
ブラックボックスからXAI (説明可能なAI) へ - LIME (Local Interpretable Model-agnostic Explanat...
 
Federated Learning
Federated LearningFederated Learning
Federated Learning
 
Classification Based Machine Learning Algorithms
Classification Based Machine Learning AlgorithmsClassification Based Machine Learning Algorithms
Classification Based Machine Learning Algorithms
 
Interpretability beyond feature attribution quantitative testing with concept...
Interpretability beyond feature attribution quantitative testing with concept...Interpretability beyond feature attribution quantitative testing with concept...
Interpretability beyond feature attribution quantitative testing with concept...
 
Machine learning
Machine learningMachine learning
Machine learning
 
Explainable AI
Explainable AIExplainable AI
Explainable AI
 
Machine Learning Unit 1 Semester 3 MSc IT Part 2 Mumbai University
Machine Learning Unit 1 Semester 3  MSc IT Part 2 Mumbai UniversityMachine Learning Unit 1 Semester 3  MSc IT Part 2 Mumbai University
Machine Learning Unit 1 Semester 3 MSc IT Part 2 Mumbai University
 
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
 
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
 
Intro to modelling-supervised learning
Intro to modelling-supervised learningIntro to modelling-supervised learning
Intro to modelling-supervised learning
 
Siamese networks
Siamese networksSiamese networks
Siamese networks
 
Introduction to ML (Machine Learning)
Introduction to ML (Machine Learning)Introduction to ML (Machine Learning)
Introduction to ML (Machine Learning)
 
eScience SHAP talk
eScience SHAP talkeScience SHAP talk
eScience SHAP talk
 
The fundamentals of Machine Learning
The fundamentals of Machine LearningThe fundamentals of Machine Learning
The fundamentals of Machine Learning
 

Similar to Introduction to Interpretable Machine Learning

Model Evaluation in the land of Deep Learning
Model Evaluation in the land of Deep LearningModel Evaluation in the land of Deep Learning
Model Evaluation in the land of Deep LearningPramit Choudhary
 
fuzzy LBP for face recognition ppt
fuzzy LBP for face recognition pptfuzzy LBP for face recognition ppt
fuzzy LBP for face recognition pptAbdullah Gubbi
 
COLOUR IMAGE PROCESSING USING MODIFIED QUATERNION NEURAL NETWORK.pdf
COLOUR IMAGE PROCESSING USING MODIFIED QUATERNION NEURAL NETWORK.pdfCOLOUR IMAGE PROCESSING USING MODIFIED QUATERNION NEURAL NETWORK.pdf
COLOUR IMAGE PROCESSING USING MODIFIED QUATERNION NEURAL NETWORK.pdfvaishaliraut8910
 
Image classification with Deep Neural Networks
Image classification with Deep Neural NetworksImage classification with Deep Neural Networks
Image classification with Deep Neural NetworksYogendra Tamang
 
abstrakty přijatých příspěvků.doc
abstrakty přijatých příspěvků.docabstrakty přijatých příspěvků.doc
abstrakty přijatých příspěvků.docbutest
 
最近の研究情勢についていくために - Deep Learningを中心に -
最近の研究情勢についていくために - Deep Learningを中心に - 最近の研究情勢についていくために - Deep Learningを中心に -
最近の研究情勢についていくために - Deep Learningを中心に - Hiroshi Fukui
 
Optimized Neural Network for Classification of Multispectral Images
Optimized Neural Network for Classification of Multispectral ImagesOptimized Neural Network for Classification of Multispectral Images
Optimized Neural Network for Classification of Multispectral ImagesIDES Editor
 
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!Sri Ambati
 
Are Human-generated Demonstrations Necessary for In-context Learning?
Are Human-generated Demonstrations Necessary for In-context Learning?Are Human-generated Demonstrations Necessary for In-context Learning?
Are Human-generated Demonstrations Necessary for In-context Learning?MENGSAYLOEM1
 
imageclassification-160206090009.pdf
imageclassification-160206090009.pdfimageclassification-160206090009.pdf
imageclassification-160206090009.pdfKammetaJoshna
 
A Parallel Framework For Multilayer Perceptron For Human Face Recognition
A Parallel Framework For Multilayer Perceptron For Human Face RecognitionA Parallel Framework For Multilayer Perceptron For Human Face Recognition
A Parallel Framework For Multilayer Perceptron For Human Face RecognitionCSCJournals
 
Image Segmentation and Classification using Neural Network
Image Segmentation and Classification using Neural NetworkImage Segmentation and Classification using Neural Network
Image Segmentation and Classification using Neural NetworkAIRCC Publishing Corporation
 
Image Segmentation and Classification using Neural Network
Image Segmentation and Classification using Neural NetworkImage Segmentation and Classification using Neural Network
Image Segmentation and Classification using Neural NetworkAIRCC Publishing Corporation
 
UTILIZING XAI TECHNIQUE TO IMPROVE AUTOENCODER BASED MODEL FOR COMPUTER NETWO...
UTILIZING XAI TECHNIQUE TO IMPROVE AUTOENCODER BASED MODEL FOR COMPUTER NETWO...UTILIZING XAI TECHNIQUE TO IMPROVE AUTOENCODER BASED MODEL FOR COMPUTER NETWO...
UTILIZING XAI TECHNIQUE TO IMPROVE AUTOENCODER BASED MODEL FOR COMPUTER NETWO...IJCNCJournal
 
Utilizing XAI Technique to Improve Autoencoder based Model for Computer Netwo...
Utilizing XAI Technique to Improve Autoencoder based Model for Computer Netwo...Utilizing XAI Technique to Improve Autoencoder based Model for Computer Netwo...
Utilizing XAI Technique to Improve Autoencoder based Model for Computer Netwo...IJCNCJournal
 
Machine learning in science and industry — day 4
Machine learning in science and industry — day 4Machine learning in science and industry — day 4
Machine learning in science and industry — day 4arogozhnikov
 
deeplearningpresentation-180625071236.pptx
deeplearningpresentation-180625071236.pptxdeeplearningpresentation-180625071236.pptx
deeplearningpresentation-180625071236.pptxJeetDesai14
 
240311_Thuy_Labseminar[Contrastive Multi-View Representation Learning on Grap...
240311_Thuy_Labseminar[Contrastive Multi-View Representation Learning on Grap...240311_Thuy_Labseminar[Contrastive Multi-View Representation Learning on Grap...
240311_Thuy_Labseminar[Contrastive Multi-View Representation Learning on Grap...thanhdowork
 

Similar to Introduction to Interpretable Machine Learning (20)

Model Evaluation in the land of Deep Learning
Model Evaluation in the land of Deep LearningModel Evaluation in the land of Deep Learning
Model Evaluation in the land of Deep Learning
 
fuzzy LBP for face recognition ppt
fuzzy LBP for face recognition pptfuzzy LBP for face recognition ppt
fuzzy LBP for face recognition ppt
 
XAI (IIT-Patna).pdf
XAI (IIT-Patna).pdfXAI (IIT-Patna).pdf
XAI (IIT-Patna).pdf
 
COLOUR IMAGE PROCESSING USING MODIFIED QUATERNION NEURAL NETWORK.pdf
COLOUR IMAGE PROCESSING USING MODIFIED QUATERNION NEURAL NETWORK.pdfCOLOUR IMAGE PROCESSING USING MODIFIED QUATERNION NEURAL NETWORK.pdf
COLOUR IMAGE PROCESSING USING MODIFIED QUATERNION NEURAL NETWORK.pdf
 
Image classification with Deep Neural Networks
Image classification with Deep Neural NetworksImage classification with Deep Neural Networks
Image classification with Deep Neural Networks
 
abstrakty přijatých příspěvků.doc
abstrakty přijatých příspěvků.docabstrakty přijatých příspěvků.doc
abstrakty přijatých příspěvků.doc
 
最近の研究情勢についていくために - Deep Learningを中心に -
最近の研究情勢についていくために - Deep Learningを中心に - 最近の研究情勢についていくために - Deep Learningを中心に -
最近の研究情勢についていくために - Deep Learningを中心に -
 
Optimized Neural Network for Classification of Multispectral Images
Optimized Neural Network for Classification of Multispectral ImagesOptimized Neural Network for Classification of Multispectral Images
Optimized Neural Network for Classification of Multispectral Images
 
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
Get hands-on with Explainable AI at Machine Learning Interpretability(MLI) Gym!
 
Are Human-generated Demonstrations Necessary for In-context Learning?
Are Human-generated Demonstrations Necessary for In-context Learning?Are Human-generated Demonstrations Necessary for In-context Learning?
Are Human-generated Demonstrations Necessary for In-context Learning?
 
imageclassification-160206090009.pdf
imageclassification-160206090009.pdfimageclassification-160206090009.pdf
imageclassification-160206090009.pdf
 
A Parallel Framework For Multilayer Perceptron For Human Face Recognition
A Parallel Framework For Multilayer Perceptron For Human Face RecognitionA Parallel Framework For Multilayer Perceptron For Human Face Recognition
A Parallel Framework For Multilayer Perceptron For Human Face Recognition
 
Image Segmentation and Classification using Neural Network
Image Segmentation and Classification using Neural NetworkImage Segmentation and Classification using Neural Network
Image Segmentation and Classification using Neural Network
 
Image Segmentation and Classification using Neural Network
Image Segmentation and Classification using Neural NetworkImage Segmentation and Classification using Neural Network
Image Segmentation and Classification using Neural Network
 
graziani_bias.pdf
graziani_bias.pdfgraziani_bias.pdf
graziani_bias.pdf
 
UTILIZING XAI TECHNIQUE TO IMPROVE AUTOENCODER BASED MODEL FOR COMPUTER NETWO...
UTILIZING XAI TECHNIQUE TO IMPROVE AUTOENCODER BASED MODEL FOR COMPUTER NETWO...UTILIZING XAI TECHNIQUE TO IMPROVE AUTOENCODER BASED MODEL FOR COMPUTER NETWO...
UTILIZING XAI TECHNIQUE TO IMPROVE AUTOENCODER BASED MODEL FOR COMPUTER NETWO...
 
Utilizing XAI Technique to Improve Autoencoder based Model for Computer Netwo...
Utilizing XAI Technique to Improve Autoencoder based Model for Computer Netwo...Utilizing XAI Technique to Improve Autoencoder based Model for Computer Netwo...
Utilizing XAI Technique to Improve Autoencoder based Model for Computer Netwo...
 
Machine learning in science and industry — day 4
Machine learning in science and industry — day 4Machine learning in science and industry — day 4
Machine learning in science and industry — day 4
 
deeplearningpresentation-180625071236.pptx
deeplearningpresentation-180625071236.pptxdeeplearningpresentation-180625071236.pptx
deeplearningpresentation-180625071236.pptx
 
240311_Thuy_Labseminar[Contrastive Multi-View Representation Learning on Grap...
240311_Thuy_Labseminar[Contrastive Multi-View Representation Learning on Grap...240311_Thuy_Labseminar[Contrastive Multi-View Representation Learning on Grap...
240311_Thuy_Labseminar[Contrastive Multi-View Representation Learning on Grap...
 

More from Nguyen Giang

Show observe and tell giang nguyen
Show observe and tell   giang nguyenShow observe and tell   giang nguyen
Show observe and tell giang nguyenNguyen Giang
 
Introduction to continual learning
Introduction to continual learningIntroduction to continual learning
Introduction to continual learningNguyen Giang
 
Variational continual learning
Variational continual learningVariational continual learning
Variational continual learningNguyen Giang
 
Survey on Script-based languages to write a Chatbot
Survey on Script-based languages to write a ChatbotSurvey on Script-based languages to write a Chatbot
Survey on Script-based languages to write a ChatbotNguyen Giang
 
How Tala works in credit score
How Tala works in credit scoreHow Tala works in credit score
How Tala works in credit scoreNguyen Giang
 
Virtual assistant with amazon alexa
Virtual assistant with amazon alexaVirtual assistant with amazon alexa
Virtual assistant with amazon alexaNguyen Giang
 
ECG Detector deployed based on OPENMSP430 open-core
ECG Detector deployed based on OPENMSP430 open-coreECG Detector deployed based on OPENMSP430 open-core
ECG Detector deployed based on OPENMSP430 open-coreNguyen Giang
 

More from Nguyen Giang (9)

Show observe and tell giang nguyen
Show observe and tell   giang nguyenShow observe and tell   giang nguyen
Show observe and tell giang nguyen
 
Introduction to continual learning
Introduction to continual learningIntroduction to continual learning
Introduction to continual learning
 
Variational continual learning
Variational continual learningVariational continual learning
Variational continual learning
 
Scalability fs v2
Scalability fs v2Scalability fs v2
Scalability fs v2
 
Survey on Script-based languages to write a Chatbot
Survey on Script-based languages to write a ChatbotSurvey on Script-based languages to write a Chatbot
Survey on Script-based languages to write a Chatbot
 
How Tala works in credit score
How Tala works in credit scoreHow Tala works in credit score
How Tala works in credit score
 
Virtual assistant with amazon alexa
Virtual assistant with amazon alexaVirtual assistant with amazon alexa
Virtual assistant with amazon alexa
 
AIML Introduction
AIML IntroductionAIML Introduction
AIML Introduction
 
ECG Detector deployed based on OPENMSP430 open-core
ECG Detector deployed based on OPENMSP430 open-coreECG Detector deployed based on OPENMSP430 open-core
ECG Detector deployed based on OPENMSP430 open-core
 

Recently uploaded

chapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineeringchapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineeringmulugeta48
 
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank  Design by Working Stress - IS Method.pdfIntze Overhead Water Tank  Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank Design by Working Stress - IS Method.pdfSuman Jyoti
 
Thermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.pptThermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.pptDineshKumar4165
 
Intro To Electric Vehicles PDF Notes.pdf
Intro To Electric Vehicles PDF Notes.pdfIntro To Electric Vehicles PDF Notes.pdf
Intro To Electric Vehicles PDF Notes.pdfrs7054576148
 
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Bookingroncy bisnoi
 
Work-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptxWork-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptxJuliansyahHarahap1
 
Unit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdfUnit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdfRagavanV2
 
Unleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leapUnleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leapRishantSharmaFr
 
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...Call Girls in Nagpur High Profile
 
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Bookingroncy bisnoi
 
UNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its PerformanceUNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its Performancesivaprakash250
 
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete RecordCCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete RecordAsst.prof M.Gokilavani
 
Double Revolving field theory-how the rotor develops torque
Double Revolving field theory-how the rotor develops torqueDouble Revolving field theory-how the rotor develops torque
Double Revolving field theory-how the rotor develops torqueBhangaleSonal
 
notes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.pptnotes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.pptMsecMca
 
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdfONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdfKamal Acharya
 
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756dollysharma2066
 
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXssuser89054b
 

Recently uploaded (20)

chapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineeringchapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineering
 
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank  Design by Working Stress - IS Method.pdfIntze Overhead Water Tank  Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
 
Thermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.pptThermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.ppt
 
Intro To Electric Vehicles PDF Notes.pdf
Intro To Electric Vehicles PDF Notes.pdfIntro To Electric Vehicles PDF Notes.pdf
Intro To Electric Vehicles PDF Notes.pdf
 
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
 
Work-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptxWork-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptx
 
Unit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdfUnit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdf
 
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
 
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
 
Unleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leapUnleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leap
 
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
 
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
 
UNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its PerformanceUNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its Performance
 
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete RecordCCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
 
Double Revolving field theory-how the rotor develops torque
Double Revolving field theory-how the rotor develops torqueDouble Revolving field theory-how the rotor develops torque
Double Revolving field theory-how the rotor develops torque
 
Call Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
 
notes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.pptnotes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.ppt
 
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdfONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
 
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
 
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
 

Introduction to Interpretable Machine Learning

  • 1. Introduction to Interpretable Machine Learning Presented by Giang Nguyen KAIST, Nov 2019
  • 2. Terminologies - Interpretable ML - Explanable AI - X – AI 2
  • 3. BIRD VIEW OVER EXPLAINABLE AI
  • 4. Deep Learning as Blackbox While powerful, deep learning models are difficult to interpret, and thus often treated as a blackbox. 4
  • 5. Interpretability Interpretation is the process of giving explanations to humans. Interpretability is not a well-defined concept 5
  • 6. Types of Interpretable Methods We can interpret the model either before building the model, when building it, or after building a model. Most interpretation methods for DNNs interpret the model after it is built. 6
  • 8. Using Inherently Interpretable Models (Sparse) linear models and decision trees are inherently interpretable. 8
  • 9. Attention Mechanisms Attention mechanisms guide deep neural networks to focus on relevant input features, which allows to interpret how the model made certain predictions. 9 [Bahdanau et al. 15] Neural Machine Translation by Jointly Learning to Align and Translate, ICLR 2015
  • 10. Limitation of Conventional Attention Mechanisms Conventional attention models may allocate attention inaccurately since they are trained in a weakly-supervised manner. The problem becomes more prominent when a task has no one-to-one mapping from inputs to the final predictions. 10
  • 11. Limitation of Conventional Attention Mechanisms This is because the conventional attention mechanisms do not consider uncertainties in the model and the input, which often leads to overconfident attention allocations. Such unreliability may lead to incorrect predictions and/or interpretations which can result in fatal consequences for safety-critical applications. 11
  • 13. Uncertainty Aware Attention (UA) 13 Multi-class classification performance on the three health records datasets
  • 14. Info-GAN 14 There are structures in the noise vectors that have meaningful and consistent effects on the output of the generator. However, there’s no systematic way to find these structures. The only thing affecting to the generator output is the noise input, so we have no idea how to modify the noise to generate expected images.
  • 15. Info-GAN 15 The idea is to provide a latent code, which has meaningful and consistent effects on the output - disentangled representation The hope is that if you keep the code the same and randomly change the noise, you get variations of the same digit.
  • 16. Info-GAN 16 c1 ∼ Cat(K = 10, p = 0.1)
  • 18. Understanding Black-Box Predictions Given a high-accuracy blackbox model and a prediction from it, can we answer why the model made a certain prediction? [Koh and Liang 17] tackles this question by training a model’s prediction through its learning algorithm and back to the training data. To formalize the impact of a training point on a prediction, they ask the counterfactual: What would happen if we did not have this training point or if its value were slightly changed? 18 [Koh and Liang 17] Understanding Black-box Predictions via Influence Functions, ICML 2017
  • 19. Interpretable Mimic Learning This framework is mainly based on knowledge distillation from Neural Networks. However, they use Gradient Boosting Trees (GBT) instead of another neural network as the student model since GBT satisfies our requirements for both learning capacity and interpretability. 19[Che et al. 2016] Z. Che, S. Purushotham, R. Khemani, and Y. Liu. Interpretable Deep Models for ICU outcome prediction, AMIA 2016. Knowledge distillation G. Hinton et al. 15
  • 20. Interpretable Mimic Learning The resulting simple model works even better than the best deep learning model – perhaps due to suppression of the overfitting. 20[Che et al. 2016] Z. Che, S. Purushotham, R. Khemani, and Y. Liu. Interpretable Deep Models for ICU outcome prediction, AMIA 2016.
  • 21. Visualizing Convolutional Neural Networks Propose Deconvolution Network (deconvnet) to inversely map the feature activations to pixel space and provide a sensitivity analysis to point out which regions of an image affect to decision making process the most. 21 [Zeiler and Fergus 14] Visualizing and Understanding Convolutional Networks, ECCV 2014
  • 22. Prediction difference analysis 22 The visualization method shows which pixels of a specific input image are evidence for or against a prediction [Zintgraf et al. 2017] Visualizing Deep Neural Network Decisions: Prediction Difference Analysis, ICLR 2017 Shown is the evidence for (red) and against (blue) the prediction. We see that the facial features of the cockatoo are most supportive for the decision, and parts of the body seem to constitute evidence against it.
  • 24. Understanding Data Through Examples [Kim et al. 16] propose to interpret the given data by providing examples that can show the full picture – majorities + minorities [Kim et al. 16] Examples are not Enough, Learn to Criticize! Criticism for Interpretability 24
  • 26. AI is data-driven, what we get is what we have 26
  • 36. 37 Over-generalization Over-generalization is consistent with evolutionary theory [Zebrowitz ‘10, Schaller’ 06] algorithms can help against over-generalization
  • 37. 38 Venn diagram of related works
  • 44. 45 Prototype-based classification • Use the learned prototypes for classification (nearest-neighbor)
  • 45. 46 Example Prototypes and Criticisms • USPS Digits Dataset Unrecognizable
  • 46. 47 Example Prototypes and Criticisms • ImageNet Dataset – 2 breeds of dog
  • 47. 48 Pilot study with human subjects Definition of interpretability: A method is interpretable if a user can correctly and efficiently predict the method’s results. Task: Assign a new data point to one of the groups using 1) all images 2) prototypes 3) prototypes and criticisms 4) small set of randomly selected images
  • 48. 49 Pilot study with human subjects
  • 50. Take-home messages 51 • There are three types of Interpretable Methods, but mostly after building models • Criticism and prototypes are equally important and are a step towards improving interpretability of complex data distributions • MMD-critic learns prototypes + criticisms that highlight aspects of data that are overlooked by prototypes.
  • 51. Discussion 52 • If we have the insight into a dataset, can we really build a better model? Human intuition is biased and not realiable!
  • 52. Gap in Interpretable ML research 53 • Limited work to explain the operation of RNNs, only CNN. Attention mechanism is not enough. Especially in multimodal network (CNN + RNN), this kind of research is more necessary
  • 53. Thank you for your attention!

Editor's Notes

  1. As a result of the success of deep learning over the past decade, many model success and even surpass human performance on classification tasks. However, it still remains secrect how deep learning models actually works. DL models are usually considered as black-box
  2. First and foremost, I would like to provide a bird view over X-ai
  3. As a result of the success of deep learning over the past decade, many model success and even surpass human performance on classification tasks. However, it still remains secrect how deep learning models actually works. DL models are usually considered as black-box
  4. To deal with this, interpretation should be given to support the operation of DL models. However, Interpretability is not a well-defined concept
  5. Generally speaking, interpretable methods are now divided into three categories: before building the model, when building it, or after building a model. However, Most interpretation methods for DNNs interpret the model after it is built.
  6. First, when building a new model, we can use/
  7. An intuitive example is to use a sparse models, which is easy to understand. In addition, decision tree support human intuition as we can know the decision at each stage.
  8. Another solution is to use attention mechanism as at each time step, we can adjust the focal point in input
  9. The next category, interpretation after building a mode, which covers almost all papers in this course.
  10. In a paper, Understanding Black-box Predictions via Influence Functions, Koh and Liang address the question: why the model made a certain prediction By training a model’s prediction through its learning algorithm and back to the training data. To formalize the impact of a training point on a prediction, they ask the counterfactual: What would happen if we did not have this training point or if its value were slightly changed?
  11. In paper Visualizing and Understanding Convolutional Networks, authors proposed to visualize learned representations in convolutional neural networks using deconvolution and maximally activating images.
  12. Another paper, mostly you know, Visualizing Deep Neural Network Decisions: Prediction Difference Analysis, they highlights areas in a given input image that provide evidence for or against a certain class.
  13. The paper I am gonna present today falls into this type of category, Interpretation Before Building a Model
  14. This paper explore data analysis through examples
  15. Now I will introduce the paper: Examples are not Enough, Learn to Criticize! Criticism for Interpretability
  16. AI community invents million of different DL models, but essentially, AI is data-driven, what we get is what we have. Its mean the model will behave based on the data we provide
  17. So, it would be nice if we know what we are having before building any models
  18. Imagine you are given a giant dataset, that contains one billion of data points. Before modeling, you wanna get a sense of what the data looks like. Of couse you don’t have time to look at all one billion images so you might do sampling from this group
  19. A lot of images look like this
  20. Another group shows that this kind of image is popular.
  21. But the problem is that protoptyes images don’t give you the full picture. There are also groups like this, and we need to look at them to get the full picture. Then the question is which group should we look? We want to look at important minorities. Others you can ignore.
  22. Like this one, animal laying on keyboard. These are small but noy ignorable
  23. Or this one. They are different from prototypes we have seen so far
  24. So you finally want to come up with an algorithm to efficiently select majorities and important minorities
  25. So this paper is about an algorithm of that kind. The idea is not only select prototypes but also important minorities. This helps human get better insights into a complex high dimensional dataset
  26. Now coming to related work of this paper
  27. Human tends to over-generalize and this cartoon suggest overgeneralization. This algorthim in this paper help us to minimiza the over-generalization via prototypes + criticisms However, examples are not enough. Relying only on examples to explain the models’ behavior can lead over-generalization and misunderstanding. Examples alone may be sufficient when the distribution of data points are ‘clean’ – in the sense that there exists a set of prototypical examples which sufficiently represent the data. However, this is rarely the case in real world data. For instance, fitting models to complex datasets often requires the use of regularization Here fitting models to complex datasets often requires the use of regularization means when training, we add regularization to generalize both prototype and criticism then we can not see the real distribution of data.
  28. There are number of methods to select prototypes but non of them focus on minorities. There are outlier detection methods that consider minorities however mostly focus on detecting abnormalities rather than representing the whole distribution.
  29. Now, we will explore how MMD-critic works
  30. So, technically speaking, this work will select prototypes generated from distribution p, and criticism from … Here, how can we measure the distance between the distribution, the authors propose to use MMD
  31. MMD is used to calculate the discrepancy between two distribution P and Q, by this witness function. However, this function is intractable; as a result, we need to approximate this function by sampling like this function. To further measure this function, authors use Bayesian model criticism and two-sample tests.
  32. Prototypes: min vi cac representative la se dung gan nhau Criticisms: max boi vi 2 distribution se la xa nhau
  33. Now jumping to experiments
  34. This paper conducts three experiments, both qualitatively and quantitatively
  35. Competitive performance with PS, thuat toan classifier su dung nearest neighbor de classify (clustering)
  36. Measure how well they did and how quickly they give back the response. Talking about speed first, people work fastest using prototypes (make sense vi so sample trong prototypes la it nhat)… Conclusion: When criticism is given together with prototypes, a human pilot study suggests that humans are better able to perform a predictive task that requires the data-distributions to be well-explained. This suggests that criticism and prototypes are a step towards improving interpretability of complex data distributions. (Nhom thu 3 perform tot nhat boi vi da biet nhom so 2 la prototype roi). Prototypes + criticisms works best suggest that human intuition works best if the dataset only have prototypes + criticisms => we can filter data to get only prototype+criticism, khi do human da co insight tot => co the build model tot hon
  37. The final part is conclusion