SlideShare a Scribd company logo
NeurIPS Meetup Japan 2021, Satoshi Hara
Explanation in ML
and Its Reliability
Satoshi Hara
Osaka University
1
NeurIPS Meetup Japan 2021
NeurIPS Meetup Japan 2021, Satoshi Hara
“Explanation” in ML
◼ Most of ML models are highly complex, or “black-box”.
◼ “Explanation in ML”: Obtain some useful information
from the model (in addition to prediction).
2
Preliminary
You are
sick.
Why?
Your XX
score is
too high.
You are
sick.
Why?
???
I don’t
know.
…
XX score is
too high.
Oh…
NeurIPS Meetup Japan 2021, Satoshi Hara
[Typical Explanation 1] Saliency Map
◼ Generate heatmaps where the model has focused on
when making predictions.
3
Preliminary
The outline of zebra
seems to be relevant.
NeurIPS Meetup Japan 2021, Satoshi Hara
[Typical Explanation 2] Similar Examples
◼ Provide some similar examples to the input of interest.
4
These images look similar.
The prediction “Lapwing” will
be correct.
Lapwing
Database
Provide some similar examples
Input
Prediction
Lapwing
Preliminary
NeurIPS Meetup Japan 2021, Satoshi Hara
History of “Explanation”
◼ History of Saliency Map
5
Dawn
2014 2016 2018 2020
2015 2017 2019
Exponential Growth of
Saliency Map Algos
Attack & Manipulation
Sanity Check
[Adebayo+,2018]
GuidedBP
[Springenberg+,2014]
DeepLIFT
[Shrikumar+,2017]
Grad-CAM
[Selvaraju+,2017]
ROAR
[Hooker+,2019]
MoRF/Deletion Metric
[Bach+,2015; Vitali+,2018]
LeRF/Insertion Metric
[Arras+,2017; Vitali+,2018]
Sensitivity
[Kindermans+,2017]
Evaluation Methods
Saliency
[Simonyan+,2014]
IntGrad
[Sundararajan+,2017]
SHAP
[Lundberg+,2017]
LIME
[Ribeiro+,2016]
LRP
[Bach+,2015]
Fairwashing
[Aivodji+,2019]
SmoothGrad
[Smilkov+,2017]
DeepTaylor
[Montavon+,2017]
Occlusion
[Zeiler+,2014]
CAM
[Zhou+,2016]
Manipulation
[Domobrowski+,2019]
The papers on “Explanation”
increased exponentially.
2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008
800
700
600
500
400
300
200
100
0
Searched
“Interpretable Machine Learning”
and
“Explainable AI”
on Web of Science
Preliminary
NeurIPS Meetup Japan 2021, Satoshi Hara
History of “Explanation”
◼ History of Saliency Map
6
Dawn
2014 2016 2018 2020
2015 2017 2019
Exponential Growth of
Saliency Map Algos
Attack & Manipulation
Sanity Check
[Adebayo+,2018]
GuidedBP
[Springenberg+,2014]
DeepLIFT
[Shrikumar+,2017]
Grad-CAM
[Selvaraju+,2017]
ROAR
[Hooker+,2019]
MoRF/Deletion Metric
[Bach+,2015; Vitali+,2018]
LeRF/Insertion Metric
[Arras+,2017; Vitali+,2018]
Sensitivity
[Kindermans+,2017]
Evaluation Methods
Saliency
[Simonyan+,2014]
IntGrad
[Sundararajan+,2017]
SHAP
[Lundberg+,2017]
LIME
[Ribeiro+,2016]
LRP
[Bach+,2015]
Fairwashing
[Aivodji+,2019]
SmoothGrad
[Smilkov+,2017]
DeepTaylor
[Montavon+,2017]
Occlusion
[Zeiler+,2014]
CAM
[Zhou+,2016]
Manipulation
[Domobrowski+,2019]
The papers on “Explanation”
increased exponentially.
800
700
600
500
400
300
200
100
0
Searched
“Interpretable Machine Learning”
and
“Explainable AI”
on Web of Science
Reliability of “Explanation” has raised
as a crucial concern.
Are the “Explanation” truly valid?
With “Explanation”, how malicious
we can be?
Preliminary
2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008
NeurIPS Meetup Japan 2021, Satoshi Hara
Technical / Social Reliability of “Explanation”
Technical Reliability “Is the explanation valid?”
What we care:
• Do the algorithms output valid “Explanation”?
Research Question:
• How can we evaluate the validity of “Explanation”?
Social Reliability “Does explanation harm the society?”
What we care:
• What will happen if we introduce “Explanation” to society?
Research Question:
• Are there any malicious use cases of “Explanation”?
7
Technical Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Faithfulness & Plausibility of “Explanation”
◼ Faithfulness [Lakkaraju+’19; Jacovi+’20]
• Does “Explanation” reflect the model’s reasoning process?
- Our interest is “How and why the model predicted that way.”
• Any “Explanation” irrelevant to the reasoning process is invalid.
- e.g. “Explanation” outputs something independent of the model.
◼ Plausibility [Lage+’19; Strout+’19]
• Does “Explanation” make sense to the users?
• Any “Explanation” unacceptable by the users is not ideal.
- e.g. Entire program code; Very noisy saliency map.
8
Technical Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Evaluation of “Explanation”
◼ Based on Faithfulness
• Sanity Checks for Saliency Maps, NeurIPS’18.
- Julius Adebayo, Justin Gilmer, Michael Muelly, Ian Goodfellow, Moritz Hardt, Been Kim
• An epoch-making paper by Google Brain.
• Evaluation of Faithfulness for saliency maps.
◼ Based on Plausibility
• Evaluation of Similarity-based Explanations, ICLR’21.
- Kazuaki Hanawa, Sho Yokoi, Satoshi Hara, Kentaro Inui
• Evaluation of Plausibility for similarity-based explanations.
9
NeurIPS Meetup Japan 2021, Satoshi Hara
Evaluation of Saliency Map
◼ Plausibility
• All the maps look more or less plausible.
• Gradient, IntegratedGrad are bit noisy.
◼ Faithfulness?
10
Technical Reliability
The outline of zebra
seems to be relevant.
NeurIPS Meetup Japan 2021, Satoshi Hara
Evaluation of Faithfulness is Not Possible.
◼ Faithfulness
• Does “Explanation” reflect the model’s reasoning process?
◼ Alternative: Sanity Check
• Check the necessary condition for faithful “Explanation”.
◼ Q. What is the necessary condition?
• “Explanation” is model-dependent.
- Any “Explanation” irrelevant to the reasoning process is invalid.
11
Unknown
→ We cannot compare with Ground Truth.
[Remark] Passing Sanity Check alone
does not guarantee faithfulness.
Technical Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Model Parameter Randomization Test
◼ Compare “Explanation” of two models with different
reasoning processes.
• Faithful “Explanation” → Outputs are different.
• Non-Faithful “Explanation” → Outputs can be identical.
12
Satisfies the necessary condition.
Passed the sanity check.
Technical Reliability
[Assumption]
These models have
different reasoning
processes.
Model 1: Fully Trained Model 2: Randomly Initialized
Input “Explanation”
by Algo. 1
“Explanation”
by Algo. 2
“Explanation” by Algo. 1 are different.
“Explanation” by Algo. 2 are identical.
Violates the necessary condition.
Failed the sanity check.
NeurIPS Meetup Japan 2021, Satoshi Hara
Model Parameter Randomization Test
◼ Model 2: DNN with last few layers randomized.
• Saliency Maps of Guided Backprop and Guided GradCAM are
invariant against model randomization.
→ They violate the necessary condition for faithfulness.
13
Model
1
Model
2
[Ref] Sanity Checks for Saliency Maps
Technical Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Evaluation of “Explanation”
◼ Based on Faithfulness
• Sanity Checks for Saliency Maps, NeurIPS’18.
- Julius Adebayo, Justin Gilmer, Michael Muelly, Ian Goodfellow, Moritz Hardt, Been Kim
• An epoch-making paper by Google Brain.
• Evaluation of Faithfulness for saliency maps.
◼ Based on Plausibility
• Evaluation of Similarity-based Explanations, ICLR’21.
- Kazuaki Hanawa, Sho Yokoi, Satoshi Hara, Kentaro Inui
• Evaluation of Plausibility for similarity-based explanations.
14
NeurIPS Meetup Japan 2021, Satoshi Hara
Evaluation of Similarity-based Explanation
◼ Faithfulness
• We can use Model Parameter Randomization Test.
◼ Plausibility?
15
These images look similar.
The prediction “Lapwing” will
be correct.
Lapwing
Database
Provide some similar examples
Input
Prediction
Lapwing
Technical Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Plausibility in Similarity-based Explanation
◼ Example
• Explanation B won’t be acceptable by the users.
- Plausibility of Explanation A > Plausibility of Explanation B
16
Database
frog
Explanation A
Database
truck
Explanation B
frog
Input
Prediction
Technical Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Evaluation of Plausibility is Not Possible.
◼ There is no universal criterion that determines the
acceptability of the users.
◼ Alternative: Sanity Check
• Check the necessary condition for faithful “Plausibility”.
◼ Q. What is the necessary condition?
• Obtained similar instance should belong to the same class.
17
is cat because a similar is cat.
is cat because a similar is dog.
Plausible
Non-Plausible
Identical Class Test
Technical Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Identical Class Test
18
Input
Dot Last Layer
All Layers
Input
Cos Last Layer
All Layers
Input
L2 Dist. Last Layer
All Layers
Influence Function
Relative IF
Fisher Kernel
Dot
Cos
Parameter Grad.
Fraction of Test Instances Passed Identical Class Test
0 0.5 1.0 0 0.5 1.0
(Image Clf.)
CIFAR10
+ CNN
(Text Clf.)
AGNews
+ Bi-LSTM
Cosine similarity of the
parameter gradient
performed almost perfectly.
Technical Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Cosine of Parameter Gradient
• GC 𝑧, 𝑧′ =
∇𝜃ℓ 𝑦,𝑓𝜃 𝑥 ,∇𝜃ℓ 𝑦′,𝑓𝜃 𝑥′
∇𝜃ℓ 𝑦,𝑓𝜃 𝑥 ∇𝜃ℓ 𝑦′,𝑓𝜃 𝑥′
19
Sussex spaniel beer bottle mobile house
Technical Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Technical / Social Reliability of “Explanation”
Technical Reliability “Is the explanation valid?”
What we care:
• Do the algorithms output valid “Explanation”?
Research Question:
• How can we evaluate the validity of “Explanation”?
Social Reliability “Does explanation harm the society?”
What we care:
• What will happen if we introduce “Explanation” to society?
Research Question:
• Are there any malicious use cases of “Explanation”?
20
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Malicious Use Cases of “Explanation”
◼ Q. Are there malicious use cases of “Explanation”?
A. Some may try to deceive people
by providing fake explanations.
◼ Q. When and why fake explanations can be used?
A. Fake explanations can show models better,
e.g., by pretending as if the models are fair.
◼ Q. Why we need to research fake explanations?
Are you evil?
A. We need to know how malicious we can be with fake
explanations. Otherwise, we cannot defend against
possible maliciousness.
21
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Fake “Explanation” for Fairness
◼ Fairness in ML
• Models can be biased towards gender, race, etc.
• Ensuring fairness of the models is crucial nowadays.
◼ What if we cannot detect the use of unfair models?
• Some may use unfair models.
- Unfair models are typically more accurate than the fair ones.
22
Social Reliability
Our model is the most accurate one in this business field.
(because of the use of unfair yet accurate model)
Moreover, our model is fair without any bias.
(by showing fake explanation)
NeurIPS Meetup Japan 2021, Satoshi Hara
Fake “Explanation” for Fairness
◼ Fake “Explanation” by Surrogate Models
• Fairwashing: the risk of rationalization, ICML’19.
- Ulrich Aïvodji, Hiromi Arai, Olivier Fortineau, Sébastien Gambs, Satoshi Hara, Alain Tapp
• Characterizing the risk of fairwashing, NeurIPS’21.
- Ulrich Aïvodji, Hiromi Arai, Sébastien Gambs, Satoshi Hara
◼ Fake “Explanation” by Examples
• Faking Fairness via Stealthily Biased Sampling, AAAI’20.
- Kazuto Fukuchi, Satoshi Hara, Takanori Maehara
◼ Ref.
• It’s Too Easy to Hide Bias in Deep-Learning Systems,
IEEE Spectrum, 2021.
23
NeurIPS Meetup Japan 2021, Satoshi Hara
The risk of “Fairwashing”
◼ Explaining fairness
24
an honest explanation
Your loan application is rejected
because your gender is …
Unfair AI: reject applicants
based on their gender.
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
The risk of “Fairwashing”
◼ Explaining fairness
25
a dishonest explanation
Your loan application is rejected
because your income is low.
Unfair AI: reject applicants
based on their gender.
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
The risk of “Fairwashing”
◼ Explaining fairness
26
Unfair AI: reject applicants
based on their gender.
a dishonest explanation
Your loan application is rejected
because your income is low.
“Fairwashing”
Malicious decision-makers can disclose a fake
explanation to rationalize their unfair decisions.
“Fairwashing”
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
The risk of “Fairwashing”
◼ Explaining fairness
27
Unfair AI: reject applicants
based on their gender.
a dishonest explanation
Your loan application is rejected
because your income is low.
This Study: LaundryML
Possible to systematically generate
fake explanations.
Raise the awareness of the risk of
“Fairwashing”.
“Fairwashing”
Malicious decision-makers can disclose a fake
explanation to rationalize their unfair decisions.
“Fairwashing”
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
◼ The idea
Generate many explanations,
and pick one that is useful for “Fairwashing”.
◼ many explanations
• Use “Model Enumeration” [Hara & Maehara’17; Hara & Ishihata’18]
• Enumerate explanation models.
◼ pick one
• Use fairness metrices such as demographic parity (DP).
• Pick an explanation most faithful to the model, with DP less
than a threshold.
28
LaundryML
Systematically generating fake explanations
The idea
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Result
◼ “Fairwashing” for decisions on Adult dataset
• Feature importance by FairML on “gender” has dropped.
29
A naïve explanation A fake explanation
gender
gender
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Result
◼ “Fairwashing” for decisions on Adult dataset
• Feature importance by FairML on “gender” has dropped.
30
A naïve explanation A false explanation
gender
gender
If
else if
else if
else if
else if
else low-income
then high-income
then low-income
then low-income
then low-income
then high-income
capital gain > 7056
marital = single
education = HS-grad
occupation = other
occupation = white-colloar
Fake Explanation
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Fake “Explanation” for Fairness
◼ Fake “Explanation” by Surrogate Models
• Fairwashing: the risk of rationalization, ICML’19.
- Ulrich Aïvodji, Hiromi Arai, Olivier Fortineau, Sébastien Gambs, Satoshi Hara, Alain Tapp
• Characterizing the risk of fairwashing, NeurIPS’21.
- Ulrich Aïvodji, Hiromi Arai, Sébastien Gambs, Satoshi Hara
◼ Fake “Explanation” by Examples
• Faking Fairness via Stealthily Biased Sampling, AAAI’20.
- Kazuto Fukuchi, Satoshi Hara, Takanori Maehara
◼ Ref.
• It’s Too Easy to Hide Bias in Deep-Learning Systems,
IEEE Spectrum, 2021.
31
NeurIPS Meetup Japan 2021, Satoshi Hara
Fairness Metrics
◼ Quantifying fairness of the models
• Several metrics + toolboxes
- FairML, AI Fairness 360 [Bellamy+’19], Aequitas [Saleiro+’18]
32
AI Fairness 360
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Fake Fairness Metrics
33
Malicious Party
Unfair Model
Service
Fairness Metric
as Evidence
Is this a fake
metric.
No guarantee whether the metrics are
computed appropriately.
Impossible to determine fake or not.
Metric alone is not a valid evidence of fairness.
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Avoiding Fake Fairness Metrics
34
Malicious Party
Unfair Model
Service
Benchmark Data
as Evidence
Fairness metric computed
on the benchmark is fair!
The metric is reproducible
using the benchmark data.
We can avoid fake!
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
(Failed) Avoiding Fake Fairness Metrics
35
Malicious Party
Unfair Model
Service
Benchmark Data
as Evidence
Fairness metric computed
on the benchmark is fair!
The metric is reproducible
using the benchmark data.
We can avoid fake!
The benchmark data can be fake.
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Generating Fake Benchmark
◼ Subsample the benchmark dataset 𝑆
from the original dataset 𝐷.
◼ “Ideal” Fake Benchmark Dataset 𝑆
• Fairness : Fairness metric computed on 𝑆 is fair.
• Stealthiness : The distribution of 𝑆 is close to 𝐷.
36
Benchmark
Fairness
Stealthiness
“Fair” Contingency Table
Original dataset
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
参照用
データ
Goodness-of-Fit Test
Generating Fake Benchmark
◼ Optimization of 𝑆 as LP (Min-Cost Flow)
min𝑆𝑊 𝑆, 𝐷 , s. t. 𝐶 𝑆 = 𝐶𝑇
◼ Detection of fake benchmark using statistical test.
• Min. Distribution diff. ≈ small detection probability
• Rejecting 𝑝 𝑆 = 𝑝(𝐷′) with KS test is probability
at most 𝑂 𝑆 𝛼 × Distribution diff.
37
Stealthiness
(Min. Distribution diff.)
Fairness
(Constraint on Contingency Table)
Reference
Data
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Undetectability of Fake Benchmark
38
Positive Cases in Contingency Table Positive Cases in Contingency Table
Fairness
Metric
(DP)
Distribution
diff.
COMPAS
Positive Cases in Contingency Table Positive Cases in Contingency Table
Fairness
Metric
(DP)
Distribution
diff.
Adult
Random Sampling
Case-Control Sampling
Proposed Sampling
Random Sampling
Case-Control Sampling
Proposed Sampling
Proposed sampling resulted to
fairer metric.
(= achieved fake fairness)
Proposed sampling attained distribution
almost identical to the original distribution.
(= undetectable)
Social Reliability
NeurIPS Meetup Japan 2021, Satoshi Hara
Technical / Social Reliability of “Explanation”
Technical Reliability “Is the explanation valid?”
What we care:
• Do the algorithms output valid “Explanation”?
Research Question:
• How can we evaluate the validity of “Explanation”?
Social Reliability “Does explanation harm the society?”
What we care:
• What will happen if we introduce “Explanation” to society?
Research Question:
• Are there any malicious use cases of “Explanation”?
39
Summary
NeurIPS Meetup Japan 2021, Satoshi Hara
Technical / Social Reliability of “Explanation”
Technical Reliability “Is the explanation valid?”
What we care:
• Do the algorithms output valid “Explanation”?
Research Question:
• How can we evaluate the validity of “Explanation”?
Social Reliability “Does explanation harm the society?”
What we care:
• What will happen if we introduce “Explanation” to society?
Research Question:
• Are there any malicious use cases of “Explanation”?
40
Summary
How can we evaluate the validity of “Explanation”?
Which evaluation is good for which “Explanation”?
When “Explanation” can be used maliciously?
Can we detect malicious use cases?

More Related Content

What's hot

【DL輪読会】ViT + Self Supervised Learningまとめ
【DL輪読会】ViT + Self Supervised Learningまとめ【DL輪読会】ViT + Self Supervised Learningまとめ
【DL輪読会】ViT + Self Supervised Learningまとめ
Deep Learning JP
 
BlackBox モデルの説明性・解釈性技術の実装
BlackBox モデルの説明性・解釈性技術の実装BlackBox モデルの説明性・解釈性技術の実装
BlackBox モデルの説明性・解釈性技術の実装
Deep Learning Lab(ディープラーニング・ラボ)
 
これからの Vision & Language ~ Acadexit した4つの理由
これからの Vision & Language ~ Acadexit した4つの理由これからの Vision & Language ~ Acadexit した4つの理由
これからの Vision & Language ~ Acadexit した4つの理由
Yoshitaka Ushiku
 
因果探索: 基本から最近の発展までを概説
因果探索: 基本から最近の発展までを概説因果探索: 基本から最近の発展までを概説
因果探索: 基本から最近の発展までを概説
Shiga University, RIKEN
 
機械学習モデルの判断根拠の説明
機械学習モデルの判断根拠の説明機械学習モデルの判断根拠の説明
機械学習モデルの判断根拠の説明
Satoshi Hara
 
Curriculum Learning (関東CV勉強会)
Curriculum Learning (関東CV勉強会)Curriculum Learning (関東CV勉強会)
Curriculum Learning (関東CV勉強会)
Yoshitaka Ushiku
 
公平性を保証したAI/機械学習
アルゴリズムの最新理論
公平性を保証したAI/機械学習
アルゴリズムの最新理論公平性を保証したAI/機械学習
アルゴリズムの最新理論
公平性を保証したAI/機械学習
アルゴリズムの最新理論
Kazuto Fukuchi
 
機械学習のためのベイズ最適化入門
機械学習のためのベイズ最適化入門機械学習のためのベイズ最適化入門
機械学習のためのベイズ最適化入門
hoxo_m
 
Sparse Codingをなるべく数式を使わず理解する(PCAやICAとの関係)
Sparse Codingをなるべく数式を使わず理解する(PCAやICAとの関係)Sparse Codingをなるべく数式を使わず理解する(PCAやICAとの関係)
Sparse Codingをなるべく数式を使わず理解する(PCAやICAとの関係)
Teppei Kurita
 
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
SSII
 
顕著性マップの推定手法
顕著性マップの推定手法顕著性マップの推定手法
顕著性マップの推定手法
Takao Yamanaka
 
近年のHierarchical Vision Transformer
近年のHierarchical Vision Transformer近年のHierarchical Vision Transformer
近年のHierarchical Vision Transformer
Yusuke Uchida
 
Dimensionality reduction with t-SNE(Rtsne) and UMAP(uwot) using R packages.
Dimensionality reduction with t-SNE(Rtsne) and UMAP(uwot) using R packages. Dimensionality reduction with t-SNE(Rtsne) and UMAP(uwot) using R packages.
Dimensionality reduction with t-SNE(Rtsne) and UMAP(uwot) using R packages.
Satoshi Kato
 
機械学習で嘘をつく話
機械学習で嘘をつく話機械学習で嘘をつく話
機械学習で嘘をつく話
Satoshi Hara
 
変分推論法(変分ベイズ法)(PRML第10章)
変分推論法(変分ベイズ法)(PRML第10章)変分推論法(変分ベイズ法)(PRML第10章)
変分推論法(変分ベイズ法)(PRML第10章)Takao Yamanaka
 
『バックドア基準の入門』@統数研研究集会
『バックドア基準の入門』@統数研研究集会『バックドア基準の入門』@統数研研究集会
『バックドア基準の入門』@統数研研究集会
takehikoihayashi
 
【DL輪読会】Toolformer: Language Models Can Teach Themselves to Use Tools
【DL輪読会】Toolformer: Language Models Can Teach Themselves to Use Tools【DL輪読会】Toolformer: Language Models Can Teach Themselves to Use Tools
【DL輪読会】Toolformer: Language Models Can Teach Themselves to Use Tools
Deep Learning JP
 
SSII2022 [OS3-02] Federated Learningの基礎と応用
SSII2022 [OS3-02] Federated Learningの基礎と応用SSII2022 [OS3-02] Federated Learningの基礎と応用
SSII2022 [OS3-02] Federated Learningの基礎と応用
SSII
 
強化学習アルゴリズムPPOの解説と実験
強化学習アルゴリズムPPOの解説と実験強化学習アルゴリズムPPOの解説と実験
強化学習アルゴリズムPPOの解説と実験
克海 納谷
 
[DL輪読会]相互情報量最大化による表現学習
[DL輪読会]相互情報量最大化による表現学習[DL輪読会]相互情報量最大化による表現学習
[DL輪読会]相互情報量最大化による表現学習
Deep Learning JP
 

What's hot (20)

【DL輪読会】ViT + Self Supervised Learningまとめ
【DL輪読会】ViT + Self Supervised Learningまとめ【DL輪読会】ViT + Self Supervised Learningまとめ
【DL輪読会】ViT + Self Supervised Learningまとめ
 
BlackBox モデルの説明性・解釈性技術の実装
BlackBox モデルの説明性・解釈性技術の実装BlackBox モデルの説明性・解釈性技術の実装
BlackBox モデルの説明性・解釈性技術の実装
 
これからの Vision & Language ~ Acadexit した4つの理由
これからの Vision & Language ~ Acadexit した4つの理由これからの Vision & Language ~ Acadexit した4つの理由
これからの Vision & Language ~ Acadexit した4つの理由
 
因果探索: 基本から最近の発展までを概説
因果探索: 基本から最近の発展までを概説因果探索: 基本から最近の発展までを概説
因果探索: 基本から最近の発展までを概説
 
機械学習モデルの判断根拠の説明
機械学習モデルの判断根拠の説明機械学習モデルの判断根拠の説明
機械学習モデルの判断根拠の説明
 
Curriculum Learning (関東CV勉強会)
Curriculum Learning (関東CV勉強会)Curriculum Learning (関東CV勉強会)
Curriculum Learning (関東CV勉強会)
 
公平性を保証したAI/機械学習
アルゴリズムの最新理論
公平性を保証したAI/機械学習
アルゴリズムの最新理論公平性を保証したAI/機械学習
アルゴリズムの最新理論
公平性を保証したAI/機械学習
アルゴリズムの最新理論
 
機械学習のためのベイズ最適化入門
機械学習のためのベイズ最適化入門機械学習のためのベイズ最適化入門
機械学習のためのベイズ最適化入門
 
Sparse Codingをなるべく数式を使わず理解する(PCAやICAとの関係)
Sparse Codingをなるべく数式を使わず理解する(PCAやICAとの関係)Sparse Codingをなるべく数式を使わず理解する(PCAやICAとの関係)
Sparse Codingをなるべく数式を使わず理解する(PCAやICAとの関係)
 
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
 
顕著性マップの推定手法
顕著性マップの推定手法顕著性マップの推定手法
顕著性マップの推定手法
 
近年のHierarchical Vision Transformer
近年のHierarchical Vision Transformer近年のHierarchical Vision Transformer
近年のHierarchical Vision Transformer
 
Dimensionality reduction with t-SNE(Rtsne) and UMAP(uwot) using R packages.
Dimensionality reduction with t-SNE(Rtsne) and UMAP(uwot) using R packages. Dimensionality reduction with t-SNE(Rtsne) and UMAP(uwot) using R packages.
Dimensionality reduction with t-SNE(Rtsne) and UMAP(uwot) using R packages.
 
機械学習で嘘をつく話
機械学習で嘘をつく話機械学習で嘘をつく話
機械学習で嘘をつく話
 
変分推論法(変分ベイズ法)(PRML第10章)
変分推論法(変分ベイズ法)(PRML第10章)変分推論法(変分ベイズ法)(PRML第10章)
変分推論法(変分ベイズ法)(PRML第10章)
 
『バックドア基準の入門』@統数研研究集会
『バックドア基準の入門』@統数研研究集会『バックドア基準の入門』@統数研研究集会
『バックドア基準の入門』@統数研研究集会
 
【DL輪読会】Toolformer: Language Models Can Teach Themselves to Use Tools
【DL輪読会】Toolformer: Language Models Can Teach Themselves to Use Tools【DL輪読会】Toolformer: Language Models Can Teach Themselves to Use Tools
【DL輪読会】Toolformer: Language Models Can Teach Themselves to Use Tools
 
SSII2022 [OS3-02] Federated Learningの基礎と応用
SSII2022 [OS3-02] Federated Learningの基礎と応用SSII2022 [OS3-02] Federated Learningの基礎と応用
SSII2022 [OS3-02] Federated Learningの基礎と応用
 
強化学習アルゴリズムPPOの解説と実験
強化学習アルゴリズムPPOの解説と実験強化学習アルゴリズムPPOの解説と実験
強化学習アルゴリズムPPOの解説と実験
 
[DL輪読会]相互情報量最大化による表現学習
[DL輪読会]相互情報量最大化による表現学習[DL輪読会]相互情報量最大化による表現学習
[DL輪読会]相互情報量最大化による表現学習
 

Similar to Explanation in Machine Learning and Its Reliability

UseR 2017
UseR 2017UseR 2017
UseR 2017
Przemek Biecek
 
Distributed processing of large graphs in python
Distributed processing of large graphs in pythonDistributed processing of large graphs in python
Distributed processing of large graphs in python
Jose Quesada (hiring)
 
All in AI: LLM Landscape & RAG in 2024 with Mark Ryan (Google) & Jerry Liu (L...
All in AI: LLM Landscape & RAG in 2024 with Mark Ryan (Google) & Jerry Liu (L...All in AI: LLM Landscape & RAG in 2024 with Mark Ryan (Google) & Jerry Liu (L...
All in AI: LLM Landscape & RAG in 2024 with Mark Ryan (Google) & Jerry Liu (L...
Daniel Zivkovic
 
DRONE: A Tool to Detect and Repair Directive Defects in Java APIs Documentation
DRONE: A Tool to Detect and Repair Directive Defects in Java APIs DocumentationDRONE: A Tool to Detect and Repair Directive Defects in Java APIs Documentation
DRONE: A Tool to Detect and Repair Directive Defects in Java APIs Documentation
Sebastiano Panichella
 
DutchMLSchool. Logistic Regression, Deepnets, Time Series
DutchMLSchool. Logistic Regression, Deepnets, Time SeriesDutchMLSchool. Logistic Regression, Deepnets, Time Series
DutchMLSchool. Logistic Regression, Deepnets, Time Series
BigML, Inc
 
Keepler Data Tech | Entendiendo tus propios modelos predictivos
Keepler Data Tech | Entendiendo tus propios modelos predictivosKeepler Data Tech | Entendiendo tus propios modelos predictivos
Keepler Data Tech | Entendiendo tus propios modelos predictivos
Keepler Data Tech
 
Workshop Tel Aviv - Graph Data Science
Workshop Tel Aviv - Graph Data ScienceWorkshop Tel Aviv - Graph Data Science
Workshop Tel Aviv - Graph Data Science
Neo4j
 
Graphs for Ai and ML
Graphs for Ai and MLGraphs for Ai and ML
Graphs for Ai and ML
Neo4j
 
Hacking Predictive Modeling - RoadSec 2018
Hacking Predictive Modeling - RoadSec 2018Hacking Predictive Modeling - RoadSec 2018
Hacking Predictive Modeling - RoadSec 2018
HJ van Veen
 
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorial
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorialBuilding Large-scale Real-world Recommender Systems - Recsys2012 tutorial
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorial
Xavier Amatriain
 
Variants of GANs - Jaejun Yoo
Variants of GANs - Jaejun YooVariants of GANs - Jaejun Yoo
Variants of GANs - Jaejun Yoo
JaeJun Yoo
 
[PR12] understanding deep learning requires rethinking generalization
[PR12] understanding deep learning requires rethinking generalization[PR12] understanding deep learning requires rethinking generalization
[PR12] understanding deep learning requires rethinking generalization
JaeJun Yoo
 
Mixed Effects Models - Random Intercepts
Mixed Effects Models - Random InterceptsMixed Effects Models - Random Intercepts
Mixed Effects Models - Random Intercepts
Scott Fraundorf
 
Neural Nets Deconstructed
Neural Nets DeconstructedNeural Nets Deconstructed
Neural Nets Deconstructed
Paul Sterk
 
Where Does It Break?
Where Does It Break?Where Does It Break?
Where Does It Break?
Frank van Harmelen
 
xai basic solutions , with some examples and formulas
xai basic solutions , with some examples and formulasxai basic solutions , with some examples and formulas
xai basic solutions , with some examples and formulas
Royi Itzhak
 
ntroducing to the Power of Graph Technology
ntroducing to the Power of Graph Technologyntroducing to the Power of Graph Technology
ntroducing to the Power of Graph Technology
Neo4j
 
Visual geometry with deep learning
Visual geometry with deep learningVisual geometry with deep learning
Visual geometry with deep learning
NAVER Engineering
 
Explainable AI
Explainable AIExplainable AI
Explainable AI
Wagston Staehler
 
Explainability and bias in AI
Explainability and bias in AIExplainability and bias in AI
Explainability and bias in AI
Bill Liu
 

Similar to Explanation in Machine Learning and Its Reliability (20)

UseR 2017
UseR 2017UseR 2017
UseR 2017
 
Distributed processing of large graphs in python
Distributed processing of large graphs in pythonDistributed processing of large graphs in python
Distributed processing of large graphs in python
 
All in AI: LLM Landscape & RAG in 2024 with Mark Ryan (Google) & Jerry Liu (L...
All in AI: LLM Landscape & RAG in 2024 with Mark Ryan (Google) & Jerry Liu (L...All in AI: LLM Landscape & RAG in 2024 with Mark Ryan (Google) & Jerry Liu (L...
All in AI: LLM Landscape & RAG in 2024 with Mark Ryan (Google) & Jerry Liu (L...
 
DRONE: A Tool to Detect and Repair Directive Defects in Java APIs Documentation
DRONE: A Tool to Detect and Repair Directive Defects in Java APIs DocumentationDRONE: A Tool to Detect and Repair Directive Defects in Java APIs Documentation
DRONE: A Tool to Detect and Repair Directive Defects in Java APIs Documentation
 
DutchMLSchool. Logistic Regression, Deepnets, Time Series
DutchMLSchool. Logistic Regression, Deepnets, Time SeriesDutchMLSchool. Logistic Regression, Deepnets, Time Series
DutchMLSchool. Logistic Regression, Deepnets, Time Series
 
Keepler Data Tech | Entendiendo tus propios modelos predictivos
Keepler Data Tech | Entendiendo tus propios modelos predictivosKeepler Data Tech | Entendiendo tus propios modelos predictivos
Keepler Data Tech | Entendiendo tus propios modelos predictivos
 
Workshop Tel Aviv - Graph Data Science
Workshop Tel Aviv - Graph Data ScienceWorkshop Tel Aviv - Graph Data Science
Workshop Tel Aviv - Graph Data Science
 
Graphs for Ai and ML
Graphs for Ai and MLGraphs for Ai and ML
Graphs for Ai and ML
 
Hacking Predictive Modeling - RoadSec 2018
Hacking Predictive Modeling - RoadSec 2018Hacking Predictive Modeling - RoadSec 2018
Hacking Predictive Modeling - RoadSec 2018
 
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorial
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorialBuilding Large-scale Real-world Recommender Systems - Recsys2012 tutorial
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorial
 
Variants of GANs - Jaejun Yoo
Variants of GANs - Jaejun YooVariants of GANs - Jaejun Yoo
Variants of GANs - Jaejun Yoo
 
[PR12] understanding deep learning requires rethinking generalization
[PR12] understanding deep learning requires rethinking generalization[PR12] understanding deep learning requires rethinking generalization
[PR12] understanding deep learning requires rethinking generalization
 
Mixed Effects Models - Random Intercepts
Mixed Effects Models - Random InterceptsMixed Effects Models - Random Intercepts
Mixed Effects Models - Random Intercepts
 
Neural Nets Deconstructed
Neural Nets DeconstructedNeural Nets Deconstructed
Neural Nets Deconstructed
 
Where Does It Break?
Where Does It Break?Where Does It Break?
Where Does It Break?
 
xai basic solutions , with some examples and formulas
xai basic solutions , with some examples and formulasxai basic solutions , with some examples and formulas
xai basic solutions , with some examples and formulas
 
ntroducing to the Power of Graph Technology
ntroducing to the Power of Graph Technologyntroducing to the Power of Graph Technology
ntroducing to the Power of Graph Technology
 
Visual geometry with deep learning
Visual geometry with deep learningVisual geometry with deep learning
Visual geometry with deep learning
 
Explainable AI
Explainable AIExplainable AI
Explainable AI
 
Explainability and bias in AI
Explainability and bias in AIExplainability and bias in AI
Explainability and bias in AI
 

More from Satoshi Hara

“機械学習の説明”の信頼性
“機械学習の説明”の信頼性“機械学習の説明”の信頼性
“機械学習の説明”の信頼性
Satoshi Hara
 
【論文調査】XAI技術の効能を ユーザ実験で評価する研究
【論文調査】XAI技術の効能を ユーザ実験で評価する研究【論文調査】XAI技術の効能を ユーザ実験で評価する研究
【論文調査】XAI技術の効能を ユーザ実験で評価する研究
Satoshi Hara
 
異常の定義と推定
異常の定義と推定異常の定義と推定
異常の定義と推定
Satoshi Hara
 
Convex Hull Approximation of Nearly Optimal Lasso Solutions
Convex Hull Approximation of Nearly Optimal Lasso SolutionsConvex Hull Approximation of Nearly Optimal Lasso Solutions
Convex Hull Approximation of Nearly Optimal Lasso Solutions
Satoshi Hara
 
Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and...
Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and...Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and...
Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and...
Satoshi Hara
 
Maximally Invariant Data Perturbation as Explanation
Maximally Invariant Data Perturbation as ExplanationMaximally Invariant Data Perturbation as Explanation
Maximally Invariant Data Perturbation as Explanation
Satoshi Hara
 
アンサンブル木モデル解釈のためのモデル簡略化法
アンサンブル木モデル解釈のためのモデル簡略化法アンサンブル木モデル解釈のためのモデル簡略化法
アンサンブル木モデル解釈のためのモデル簡略化法
Satoshi Hara
 
機械学習モデルの列挙
機械学習モデルの列挙機械学習モデルの列挙
機械学習モデルの列挙
Satoshi Hara
 
KDD'17読み会:Anomaly Detection with Robust Deep Autoencoders
KDD'17読み会:Anomaly Detection with Robust Deep AutoencodersKDD'17読み会:Anomaly Detection with Robust Deep Autoencoders
KDD'17読み会:Anomaly Detection with Robust Deep Autoencoders
Satoshi Hara
 
特徴選択のためのLasso解列挙
特徴選択のためのLasso解列挙特徴選択のためのLasso解列挙
特徴選択のためのLasso解列挙
Satoshi Hara
 

More from Satoshi Hara (10)

“機械学習の説明”の信頼性
“機械学習の説明”の信頼性“機械学習の説明”の信頼性
“機械学習の説明”の信頼性
 
【論文調査】XAI技術の効能を ユーザ実験で評価する研究
【論文調査】XAI技術の効能を ユーザ実験で評価する研究【論文調査】XAI技術の効能を ユーザ実験で評価する研究
【論文調査】XAI技術の効能を ユーザ実験で評価する研究
 
異常の定義と推定
異常の定義と推定異常の定義と推定
異常の定義と推定
 
Convex Hull Approximation of Nearly Optimal Lasso Solutions
Convex Hull Approximation of Nearly Optimal Lasso SolutionsConvex Hull Approximation of Nearly Optimal Lasso Solutions
Convex Hull Approximation of Nearly Optimal Lasso Solutions
 
Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and...
Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and...Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and...
Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and...
 
Maximally Invariant Data Perturbation as Explanation
Maximally Invariant Data Perturbation as ExplanationMaximally Invariant Data Perturbation as Explanation
Maximally Invariant Data Perturbation as Explanation
 
アンサンブル木モデル解釈のためのモデル簡略化法
アンサンブル木モデル解釈のためのモデル簡略化法アンサンブル木モデル解釈のためのモデル簡略化法
アンサンブル木モデル解釈のためのモデル簡略化法
 
機械学習モデルの列挙
機械学習モデルの列挙機械学習モデルの列挙
機械学習モデルの列挙
 
KDD'17読み会:Anomaly Detection with Robust Deep Autoencoders
KDD'17読み会:Anomaly Detection with Robust Deep AutoencodersKDD'17読み会:Anomaly Detection with Robust Deep Autoencoders
KDD'17読み会:Anomaly Detection with Robust Deep Autoencoders
 
特徴選択のためのLasso解列挙
特徴選択のためのLasso解列挙特徴選択のためのLasso解列挙
特徴選択のためのLasso解列挙
 

Recently uploaded

Transcript: Selling digital books in 2024: Insights from industry leaders - T...
Transcript: Selling digital books in 2024: Insights from industry leaders - T...Transcript: Selling digital books in 2024: Insights from industry leaders - T...
Transcript: Selling digital books in 2024: Insights from industry leaders - T...
BookNet Canada
 
Pushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 daysPushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 days
Adtran
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance
 
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdfSAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
Peter Spielvogel
 
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
Sri Ambati
 
zkStudyClub - Reef: Fast Succinct Non-Interactive Zero-Knowledge Regex Proofs
zkStudyClub - Reef: Fast Succinct Non-Interactive Zero-Knowledge Regex ProofszkStudyClub - Reef: Fast Succinct Non-Interactive Zero-Knowledge Regex Proofs
zkStudyClub - Reef: Fast Succinct Non-Interactive Zero-Knowledge Regex Proofs
Alex Pruden
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Aggregage
 
Le nuove frontiere dell'AI nell'RPA con UiPath Autopilot™
Le nuove frontiere dell'AI nell'RPA con UiPath Autopilot™Le nuove frontiere dell'AI nell'RPA con UiPath Autopilot™
Le nuove frontiere dell'AI nell'RPA con UiPath Autopilot™
UiPathCommunity
 
DevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA ConnectDevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA Connect
Kari Kakkonen
 
RESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for studentsRESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for students
KAMESHS29
 
The Future of Platform Engineering
The Future of Platform EngineeringThe Future of Platform Engineering
The Future of Platform Engineering
Jemma Hussein Allen
 
Free Complete Python - A step towards Data Science
Free Complete Python - A step towards Data ScienceFree Complete Python - A step towards Data Science
Free Complete Python - A step towards Data Science
RinaMondal9
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
KatiaHIMEUR1
 
The Metaverse and AI: how can decision-makers harness the Metaverse for their...
The Metaverse and AI: how can decision-makers harness the Metaverse for their...The Metaverse and AI: how can decision-makers harness the Metaverse for their...
The Metaverse and AI: how can decision-makers harness the Metaverse for their...
Jen Stirrup
 
Introduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - CybersecurityIntroduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - Cybersecurity
mikeeftimakis1
 
Monitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR EventsMonitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR Events
Ana-Maria Mihalceanu
 
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdfFIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance
 
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
James Anderson
 
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdfFIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Albert Hoitingh
 

Recently uploaded (20)

Transcript: Selling digital books in 2024: Insights from industry leaders - T...
Transcript: Selling digital books in 2024: Insights from industry leaders - T...Transcript: Selling digital books in 2024: Insights from industry leaders - T...
Transcript: Selling digital books in 2024: Insights from industry leaders - T...
 
Pushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 daysPushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 days
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
 
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdfSAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
SAP Sapphire 2024 - ASUG301 building better apps with SAP Fiori.pdf
 
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
 
zkStudyClub - Reef: Fast Succinct Non-Interactive Zero-Knowledge Regex Proofs
zkStudyClub - Reef: Fast Succinct Non-Interactive Zero-Knowledge Regex ProofszkStudyClub - Reef: Fast Succinct Non-Interactive Zero-Knowledge Regex Proofs
zkStudyClub - Reef: Fast Succinct Non-Interactive Zero-Knowledge Regex Proofs
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
 
Le nuove frontiere dell'AI nell'RPA con UiPath Autopilot™
Le nuove frontiere dell'AI nell'RPA con UiPath Autopilot™Le nuove frontiere dell'AI nell'RPA con UiPath Autopilot™
Le nuove frontiere dell'AI nell'RPA con UiPath Autopilot™
 
DevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA ConnectDevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA Connect
 
RESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for studentsRESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for students
 
The Future of Platform Engineering
The Future of Platform EngineeringThe Future of Platform Engineering
The Future of Platform Engineering
 
Free Complete Python - A step towards Data Science
Free Complete Python - A step towards Data ScienceFree Complete Python - A step towards Data Science
Free Complete Python - A step towards Data Science
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
 
The Metaverse and AI: how can decision-makers harness the Metaverse for their...
The Metaverse and AI: how can decision-makers harness the Metaverse for their...The Metaverse and AI: how can decision-makers harness the Metaverse for their...
The Metaverse and AI: how can decision-makers harness the Metaverse for their...
 
Introduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - CybersecurityIntroduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - Cybersecurity
 
Monitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR EventsMonitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR Events
 
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdfFIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
 
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...
 
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdfFIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
FIDO Alliance Osaka Seminar: The WebAuthn API and Discoverable Credentials.pdf
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
 

Explanation in Machine Learning and Its Reliability

  • 1. NeurIPS Meetup Japan 2021, Satoshi Hara Explanation in ML and Its Reliability Satoshi Hara Osaka University 1 NeurIPS Meetup Japan 2021
  • 2. NeurIPS Meetup Japan 2021, Satoshi Hara “Explanation” in ML ◼ Most of ML models are highly complex, or “black-box”. ◼ “Explanation in ML”: Obtain some useful information from the model (in addition to prediction). 2 Preliminary You are sick. Why? Your XX score is too high. You are sick. Why? ??? I don’t know. … XX score is too high. Oh…
  • 3. NeurIPS Meetup Japan 2021, Satoshi Hara [Typical Explanation 1] Saliency Map ◼ Generate heatmaps where the model has focused on when making predictions. 3 Preliminary The outline of zebra seems to be relevant.
  • 4. NeurIPS Meetup Japan 2021, Satoshi Hara [Typical Explanation 2] Similar Examples ◼ Provide some similar examples to the input of interest. 4 These images look similar. The prediction “Lapwing” will be correct. Lapwing Database Provide some similar examples Input Prediction Lapwing Preliminary
  • 5. NeurIPS Meetup Japan 2021, Satoshi Hara History of “Explanation” ◼ History of Saliency Map 5 Dawn 2014 2016 2018 2020 2015 2017 2019 Exponential Growth of Saliency Map Algos Attack & Manipulation Sanity Check [Adebayo+,2018] GuidedBP [Springenberg+,2014] DeepLIFT [Shrikumar+,2017] Grad-CAM [Selvaraju+,2017] ROAR [Hooker+,2019] MoRF/Deletion Metric [Bach+,2015; Vitali+,2018] LeRF/Insertion Metric [Arras+,2017; Vitali+,2018] Sensitivity [Kindermans+,2017] Evaluation Methods Saliency [Simonyan+,2014] IntGrad [Sundararajan+,2017] SHAP [Lundberg+,2017] LIME [Ribeiro+,2016] LRP [Bach+,2015] Fairwashing [Aivodji+,2019] SmoothGrad [Smilkov+,2017] DeepTaylor [Montavon+,2017] Occlusion [Zeiler+,2014] CAM [Zhou+,2016] Manipulation [Domobrowski+,2019] The papers on “Explanation” increased exponentially. 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 800 700 600 500 400 300 200 100 0 Searched “Interpretable Machine Learning” and “Explainable AI” on Web of Science Preliminary
  • 6. NeurIPS Meetup Japan 2021, Satoshi Hara History of “Explanation” ◼ History of Saliency Map 6 Dawn 2014 2016 2018 2020 2015 2017 2019 Exponential Growth of Saliency Map Algos Attack & Manipulation Sanity Check [Adebayo+,2018] GuidedBP [Springenberg+,2014] DeepLIFT [Shrikumar+,2017] Grad-CAM [Selvaraju+,2017] ROAR [Hooker+,2019] MoRF/Deletion Metric [Bach+,2015; Vitali+,2018] LeRF/Insertion Metric [Arras+,2017; Vitali+,2018] Sensitivity [Kindermans+,2017] Evaluation Methods Saliency [Simonyan+,2014] IntGrad [Sundararajan+,2017] SHAP [Lundberg+,2017] LIME [Ribeiro+,2016] LRP [Bach+,2015] Fairwashing [Aivodji+,2019] SmoothGrad [Smilkov+,2017] DeepTaylor [Montavon+,2017] Occlusion [Zeiler+,2014] CAM [Zhou+,2016] Manipulation [Domobrowski+,2019] The papers on “Explanation” increased exponentially. 800 700 600 500 400 300 200 100 0 Searched “Interpretable Machine Learning” and “Explainable AI” on Web of Science Reliability of “Explanation” has raised as a crucial concern. Are the “Explanation” truly valid? With “Explanation”, how malicious we can be? Preliminary 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008
  • 7. NeurIPS Meetup Japan 2021, Satoshi Hara Technical / Social Reliability of “Explanation” Technical Reliability “Is the explanation valid?” What we care: • Do the algorithms output valid “Explanation”? Research Question: • How can we evaluate the validity of “Explanation”? Social Reliability “Does explanation harm the society?” What we care: • What will happen if we introduce “Explanation” to society? Research Question: • Are there any malicious use cases of “Explanation”? 7 Technical Reliability
  • 8. NeurIPS Meetup Japan 2021, Satoshi Hara Faithfulness & Plausibility of “Explanation” ◼ Faithfulness [Lakkaraju+’19; Jacovi+’20] • Does “Explanation” reflect the model’s reasoning process? - Our interest is “How and why the model predicted that way.” • Any “Explanation” irrelevant to the reasoning process is invalid. - e.g. “Explanation” outputs something independent of the model. ◼ Plausibility [Lage+’19; Strout+’19] • Does “Explanation” make sense to the users? • Any “Explanation” unacceptable by the users is not ideal. - e.g. Entire program code; Very noisy saliency map. 8 Technical Reliability
  • 9. NeurIPS Meetup Japan 2021, Satoshi Hara Evaluation of “Explanation” ◼ Based on Faithfulness • Sanity Checks for Saliency Maps, NeurIPS’18. - Julius Adebayo, Justin Gilmer, Michael Muelly, Ian Goodfellow, Moritz Hardt, Been Kim • An epoch-making paper by Google Brain. • Evaluation of Faithfulness for saliency maps. ◼ Based on Plausibility • Evaluation of Similarity-based Explanations, ICLR’21. - Kazuaki Hanawa, Sho Yokoi, Satoshi Hara, Kentaro Inui • Evaluation of Plausibility for similarity-based explanations. 9
  • 10. NeurIPS Meetup Japan 2021, Satoshi Hara Evaluation of Saliency Map ◼ Plausibility • All the maps look more or less plausible. • Gradient, IntegratedGrad are bit noisy. ◼ Faithfulness? 10 Technical Reliability The outline of zebra seems to be relevant.
  • 11. NeurIPS Meetup Japan 2021, Satoshi Hara Evaluation of Faithfulness is Not Possible. ◼ Faithfulness • Does “Explanation” reflect the model’s reasoning process? ◼ Alternative: Sanity Check • Check the necessary condition for faithful “Explanation”. ◼ Q. What is the necessary condition? • “Explanation” is model-dependent. - Any “Explanation” irrelevant to the reasoning process is invalid. 11 Unknown → We cannot compare with Ground Truth. [Remark] Passing Sanity Check alone does not guarantee faithfulness. Technical Reliability
  • 12. NeurIPS Meetup Japan 2021, Satoshi Hara Model Parameter Randomization Test ◼ Compare “Explanation” of two models with different reasoning processes. • Faithful “Explanation” → Outputs are different. • Non-Faithful “Explanation” → Outputs can be identical. 12 Satisfies the necessary condition. Passed the sanity check. Technical Reliability [Assumption] These models have different reasoning processes. Model 1: Fully Trained Model 2: Randomly Initialized Input “Explanation” by Algo. 1 “Explanation” by Algo. 2 “Explanation” by Algo. 1 are different. “Explanation” by Algo. 2 are identical. Violates the necessary condition. Failed the sanity check.
  • 13. NeurIPS Meetup Japan 2021, Satoshi Hara Model Parameter Randomization Test ◼ Model 2: DNN with last few layers randomized. • Saliency Maps of Guided Backprop and Guided GradCAM are invariant against model randomization. → They violate the necessary condition for faithfulness. 13 Model 1 Model 2 [Ref] Sanity Checks for Saliency Maps Technical Reliability
  • 14. NeurIPS Meetup Japan 2021, Satoshi Hara Evaluation of “Explanation” ◼ Based on Faithfulness • Sanity Checks for Saliency Maps, NeurIPS’18. - Julius Adebayo, Justin Gilmer, Michael Muelly, Ian Goodfellow, Moritz Hardt, Been Kim • An epoch-making paper by Google Brain. • Evaluation of Faithfulness for saliency maps. ◼ Based on Plausibility • Evaluation of Similarity-based Explanations, ICLR’21. - Kazuaki Hanawa, Sho Yokoi, Satoshi Hara, Kentaro Inui • Evaluation of Plausibility for similarity-based explanations. 14
  • 15. NeurIPS Meetup Japan 2021, Satoshi Hara Evaluation of Similarity-based Explanation ◼ Faithfulness • We can use Model Parameter Randomization Test. ◼ Plausibility? 15 These images look similar. The prediction “Lapwing” will be correct. Lapwing Database Provide some similar examples Input Prediction Lapwing Technical Reliability
  • 16. NeurIPS Meetup Japan 2021, Satoshi Hara Plausibility in Similarity-based Explanation ◼ Example • Explanation B won’t be acceptable by the users. - Plausibility of Explanation A > Plausibility of Explanation B 16 Database frog Explanation A Database truck Explanation B frog Input Prediction Technical Reliability
  • 17. NeurIPS Meetup Japan 2021, Satoshi Hara Evaluation of Plausibility is Not Possible. ◼ There is no universal criterion that determines the acceptability of the users. ◼ Alternative: Sanity Check • Check the necessary condition for faithful “Plausibility”. ◼ Q. What is the necessary condition? • Obtained similar instance should belong to the same class. 17 is cat because a similar is cat. is cat because a similar is dog. Plausible Non-Plausible Identical Class Test Technical Reliability
  • 18. NeurIPS Meetup Japan 2021, Satoshi Hara Identical Class Test 18 Input Dot Last Layer All Layers Input Cos Last Layer All Layers Input L2 Dist. Last Layer All Layers Influence Function Relative IF Fisher Kernel Dot Cos Parameter Grad. Fraction of Test Instances Passed Identical Class Test 0 0.5 1.0 0 0.5 1.0 (Image Clf.) CIFAR10 + CNN (Text Clf.) AGNews + Bi-LSTM Cosine similarity of the parameter gradient performed almost perfectly. Technical Reliability
  • 19. NeurIPS Meetup Japan 2021, Satoshi Hara Cosine of Parameter Gradient • GC 𝑧, 𝑧′ = ∇𝜃ℓ 𝑦,𝑓𝜃 𝑥 ,∇𝜃ℓ 𝑦′,𝑓𝜃 𝑥′ ∇𝜃ℓ 𝑦,𝑓𝜃 𝑥 ∇𝜃ℓ 𝑦′,𝑓𝜃 𝑥′ 19 Sussex spaniel beer bottle mobile house Technical Reliability
  • 20. NeurIPS Meetup Japan 2021, Satoshi Hara Technical / Social Reliability of “Explanation” Technical Reliability “Is the explanation valid?” What we care: • Do the algorithms output valid “Explanation”? Research Question: • How can we evaluate the validity of “Explanation”? Social Reliability “Does explanation harm the society?” What we care: • What will happen if we introduce “Explanation” to society? Research Question: • Are there any malicious use cases of “Explanation”? 20 Social Reliability
  • 21. NeurIPS Meetup Japan 2021, Satoshi Hara Malicious Use Cases of “Explanation” ◼ Q. Are there malicious use cases of “Explanation”? A. Some may try to deceive people by providing fake explanations. ◼ Q. When and why fake explanations can be used? A. Fake explanations can show models better, e.g., by pretending as if the models are fair. ◼ Q. Why we need to research fake explanations? Are you evil? A. We need to know how malicious we can be with fake explanations. Otherwise, we cannot defend against possible maliciousness. 21 Social Reliability
  • 22. NeurIPS Meetup Japan 2021, Satoshi Hara Fake “Explanation” for Fairness ◼ Fairness in ML • Models can be biased towards gender, race, etc. • Ensuring fairness of the models is crucial nowadays. ◼ What if we cannot detect the use of unfair models? • Some may use unfair models. - Unfair models are typically more accurate than the fair ones. 22 Social Reliability Our model is the most accurate one in this business field. (because of the use of unfair yet accurate model) Moreover, our model is fair without any bias. (by showing fake explanation)
  • 23. NeurIPS Meetup Japan 2021, Satoshi Hara Fake “Explanation” for Fairness ◼ Fake “Explanation” by Surrogate Models • Fairwashing: the risk of rationalization, ICML’19. - Ulrich Aïvodji, Hiromi Arai, Olivier Fortineau, Sébastien Gambs, Satoshi Hara, Alain Tapp • Characterizing the risk of fairwashing, NeurIPS’21. - Ulrich Aïvodji, Hiromi Arai, Sébastien Gambs, Satoshi Hara ◼ Fake “Explanation” by Examples • Faking Fairness via Stealthily Biased Sampling, AAAI’20. - Kazuto Fukuchi, Satoshi Hara, Takanori Maehara ◼ Ref. • It’s Too Easy to Hide Bias in Deep-Learning Systems, IEEE Spectrum, 2021. 23
  • 24. NeurIPS Meetup Japan 2021, Satoshi Hara The risk of “Fairwashing” ◼ Explaining fairness 24 an honest explanation Your loan application is rejected because your gender is … Unfair AI: reject applicants based on their gender. Social Reliability
  • 25. NeurIPS Meetup Japan 2021, Satoshi Hara The risk of “Fairwashing” ◼ Explaining fairness 25 a dishonest explanation Your loan application is rejected because your income is low. Unfair AI: reject applicants based on their gender. Social Reliability
  • 26. NeurIPS Meetup Japan 2021, Satoshi Hara The risk of “Fairwashing” ◼ Explaining fairness 26 Unfair AI: reject applicants based on their gender. a dishonest explanation Your loan application is rejected because your income is low. “Fairwashing” Malicious decision-makers can disclose a fake explanation to rationalize their unfair decisions. “Fairwashing” Social Reliability
  • 27. NeurIPS Meetup Japan 2021, Satoshi Hara The risk of “Fairwashing” ◼ Explaining fairness 27 Unfair AI: reject applicants based on their gender. a dishonest explanation Your loan application is rejected because your income is low. This Study: LaundryML Possible to systematically generate fake explanations. Raise the awareness of the risk of “Fairwashing”. “Fairwashing” Malicious decision-makers can disclose a fake explanation to rationalize their unfair decisions. “Fairwashing” Social Reliability
  • 28. NeurIPS Meetup Japan 2021, Satoshi Hara ◼ The idea Generate many explanations, and pick one that is useful for “Fairwashing”. ◼ many explanations • Use “Model Enumeration” [Hara & Maehara’17; Hara & Ishihata’18] • Enumerate explanation models. ◼ pick one • Use fairness metrices such as demographic parity (DP). • Pick an explanation most faithful to the model, with DP less than a threshold. 28 LaundryML Systematically generating fake explanations The idea Social Reliability
  • 29. NeurIPS Meetup Japan 2021, Satoshi Hara Result ◼ “Fairwashing” for decisions on Adult dataset • Feature importance by FairML on “gender” has dropped. 29 A naïve explanation A fake explanation gender gender Social Reliability
  • 30. NeurIPS Meetup Japan 2021, Satoshi Hara Result ◼ “Fairwashing” for decisions on Adult dataset • Feature importance by FairML on “gender” has dropped. 30 A naïve explanation A false explanation gender gender If else if else if else if else if else low-income then high-income then low-income then low-income then low-income then high-income capital gain > 7056 marital = single education = HS-grad occupation = other occupation = white-colloar Fake Explanation Social Reliability
  • 31. NeurIPS Meetup Japan 2021, Satoshi Hara Fake “Explanation” for Fairness ◼ Fake “Explanation” by Surrogate Models • Fairwashing: the risk of rationalization, ICML’19. - Ulrich Aïvodji, Hiromi Arai, Olivier Fortineau, Sébastien Gambs, Satoshi Hara, Alain Tapp • Characterizing the risk of fairwashing, NeurIPS’21. - Ulrich Aïvodji, Hiromi Arai, Sébastien Gambs, Satoshi Hara ◼ Fake “Explanation” by Examples • Faking Fairness via Stealthily Biased Sampling, AAAI’20. - Kazuto Fukuchi, Satoshi Hara, Takanori Maehara ◼ Ref. • It’s Too Easy to Hide Bias in Deep-Learning Systems, IEEE Spectrum, 2021. 31
  • 32. NeurIPS Meetup Japan 2021, Satoshi Hara Fairness Metrics ◼ Quantifying fairness of the models • Several metrics + toolboxes - FairML, AI Fairness 360 [Bellamy+’19], Aequitas [Saleiro+’18] 32 AI Fairness 360 Social Reliability
  • 33. NeurIPS Meetup Japan 2021, Satoshi Hara Fake Fairness Metrics 33 Malicious Party Unfair Model Service Fairness Metric as Evidence Is this a fake metric. No guarantee whether the metrics are computed appropriately. Impossible to determine fake or not. Metric alone is not a valid evidence of fairness. Social Reliability
  • 34. NeurIPS Meetup Japan 2021, Satoshi Hara Avoiding Fake Fairness Metrics 34 Malicious Party Unfair Model Service Benchmark Data as Evidence Fairness metric computed on the benchmark is fair! The metric is reproducible using the benchmark data. We can avoid fake! Social Reliability
  • 35. NeurIPS Meetup Japan 2021, Satoshi Hara (Failed) Avoiding Fake Fairness Metrics 35 Malicious Party Unfair Model Service Benchmark Data as Evidence Fairness metric computed on the benchmark is fair! The metric is reproducible using the benchmark data. We can avoid fake! The benchmark data can be fake. Social Reliability
  • 36. NeurIPS Meetup Japan 2021, Satoshi Hara Generating Fake Benchmark ◼ Subsample the benchmark dataset 𝑆 from the original dataset 𝐷. ◼ “Ideal” Fake Benchmark Dataset 𝑆 • Fairness : Fairness metric computed on 𝑆 is fair. • Stealthiness : The distribution of 𝑆 is close to 𝐷. 36 Benchmark Fairness Stealthiness “Fair” Contingency Table Original dataset Social Reliability
  • 37. NeurIPS Meetup Japan 2021, Satoshi Hara 参照用 データ Goodness-of-Fit Test Generating Fake Benchmark ◼ Optimization of 𝑆 as LP (Min-Cost Flow) min𝑆𝑊 𝑆, 𝐷 , s. t. 𝐶 𝑆 = 𝐶𝑇 ◼ Detection of fake benchmark using statistical test. • Min. Distribution diff. ≈ small detection probability • Rejecting 𝑝 𝑆 = 𝑝(𝐷′) with KS test is probability at most 𝑂 𝑆 𝛼 × Distribution diff. 37 Stealthiness (Min. Distribution diff.) Fairness (Constraint on Contingency Table) Reference Data Social Reliability
  • 38. NeurIPS Meetup Japan 2021, Satoshi Hara Undetectability of Fake Benchmark 38 Positive Cases in Contingency Table Positive Cases in Contingency Table Fairness Metric (DP) Distribution diff. COMPAS Positive Cases in Contingency Table Positive Cases in Contingency Table Fairness Metric (DP) Distribution diff. Adult Random Sampling Case-Control Sampling Proposed Sampling Random Sampling Case-Control Sampling Proposed Sampling Proposed sampling resulted to fairer metric. (= achieved fake fairness) Proposed sampling attained distribution almost identical to the original distribution. (= undetectable) Social Reliability
  • 39. NeurIPS Meetup Japan 2021, Satoshi Hara Technical / Social Reliability of “Explanation” Technical Reliability “Is the explanation valid?” What we care: • Do the algorithms output valid “Explanation”? Research Question: • How can we evaluate the validity of “Explanation”? Social Reliability “Does explanation harm the society?” What we care: • What will happen if we introduce “Explanation” to society? Research Question: • Are there any malicious use cases of “Explanation”? 39 Summary
  • 40. NeurIPS Meetup Japan 2021, Satoshi Hara Technical / Social Reliability of “Explanation” Technical Reliability “Is the explanation valid?” What we care: • Do the algorithms output valid “Explanation”? Research Question: • How can we evaluate the validity of “Explanation”? Social Reliability “Does explanation harm the society?” What we care: • What will happen if we introduce “Explanation” to society? Research Question: • Are there any malicious use cases of “Explanation”? 40 Summary How can we evaluate the validity of “Explanation”? Which evaluation is good for which “Explanation”? When “Explanation” can be used maliciously? Can we detect malicious use cases?