SlideShare a Scribd company logo
1 of 110
Download to read offline
1
20 2018/12/21
n
• ~2013.3, PhD@ ,
• 2013.4~2016.3, @IBM
• 2016.4~2017.8, @ ERATO, NII
• 2017.9~, @ ,
n
•
ECML’11
AISTATS’15,17
•
AISTATS’18
AAAI’17,18
ongoing
2
•
•
•
• AISTATS’18
• AAAI’17,18
3
AI
n
•
n
•
4
…
AI
n
•
n
•
5
XX
XX
AI
n
• AI
AI
AI AI
n
• AI
•
n
6
n
•
•
• AI
•
•
7
8
EU GDPR
n GDPR-22
1. The data subject shall have the right not to be subject to a decision based
solely on automated processing, including profiling, which produces legal
effects concerning him or her or similarly significantly affects him or her.
2. Paragraph 1 shall not apply if the decision: is necessary for entering into, or
performance of, a contract between the data subject and a data controller; is
authorised by Union or Member State law to which the controller is subject
and which also lays down suitable measures to safeguard the data subject’s
rights and freedoms and legitimate interests; or is based on the data subject’s
explicit consent.
3. In the cases referred to in points (a) and (c) of paragraph 2, the data controller
shall implement suitable measures to safeguard the data subject’s rights and
freedoms and legitimate interests, at least the right to obtain human
intervention on the part of the controller, to express his or her point of view
and to contest the decision.
4. Decisions referred to in paragraph 2 shall not be based on special categories of
personal data referred to in Article 9(2)1), unless point (a) or (g) of Article 9(2)
applies and suitable measures to safeguard the data subject’s rights and
freedoms and legitimate interests are in place.
9
n 2016
• ICML, NIPS
n
• , ,
Vol.33, No.3, pages 366--369, 2018.
• , Qiita
10
n AI
11
Peeking inside the black-box: A survey on Explainable Artificial Intelligence (XAI)
https://ieeexplore.ieee.org/document/8466590/
7 SCOPUS,
IEEExplore, ACM Digital Library, Google
Scholar, Citeseer Library, ScienceDirect,
arXiv
“intelligible”,
“interpretable”, “transparency”, “black box”,
“understandable”, “comprehensible”,
“explainable” AI
“Artificial Intelligence”, “Intelligent
system”, “Machine learning”, “deep learning”,
“classifier” , “decision tree”
•
•
•
• AISTATS’18
• AAAI’17,18
12
n
•
n
• 1.
• 2.
• 3.
• 4.
• …
13
n
n /
•
n
•
•
n “ ”
•
14
1.
•
2.
•
3.
•
15
n
n Why Should I Trust You?: Explaining the Predictions of
Any Classifier, KDD'16 [Python LIME; R LIME]
n A Unified Approach to Interpreting Model Predictions,
NIPS'17 [Python SHAP]
n Anchors: High-Precision Model-Agnostic Explanations,
AAAI'18 [Python Anchor]
n Understanding Black-box Predictions via Influence
Functions, ICML’17 [Python influence-release]
16
n
n Born Again Trees
n Making Tree Ensembles Interpretable: A Bayesian Model
Selection Approach, AISTATS'18 [Python defragTrees]
17
n
n [Python+Tensorflow
saliency; DeepExplain]
• Striving for Simplicity: The All Convolutional Net
(GuidedBackprop)
• On Pixel-Wise Explanations for Non-Linear Classifier Decisions
by Layer-Wise Relevance Propagation (Epsilon-LRP)
• Axiomatic Attribution for Deep Networks (IntegratedGrad)
• SmoothGrad: Removing Noise by Adding Noise (SmoothGrad)
• Learning Important Features Through Propagating Activation
Differences (DeepLIFT)
18
19
n
n Why Should I Trust You?: Explaining the Predictions of
Any Classifier, KDD'16 [Python LIME; R LIME]
n A Unified Approach to Interpreting Model Predictions,
NIPS'17 [Python SHAP]
n Anchors: High-Precision Model-Agnostic Explanations,
AAAI'18 [Python Anchor]
n Understanding Black-box Predictions via Influence
Functions, ICML’17 [Python influence-release]
20
n
n
• LIME, SHAP, Anchor
n
• influence
21
LIME
n Why Should I Trust You?: Explaining the Predictions of
Any Classifier, KDD'16 [Python LIME; R LIME]
•
•
22
LIME
23
LIME
n 2
• One-Hot
•
n LIME
• Adult One-Hot
24
LIME
n ! !′
• LIME !′ !# ∈ {0, 1}* +
!,
#
+
n -(!) !0
• - ! ≈ 2 !# ≔ 40 + 46!# for !# ∈ NeighborOf(!0
#
)
• 4 4,
25
LIME
n !(#) #%
• ! # ≈ ' #( ≔ *% + *,#( for #( ∈ NeighborOf(#%
(
)
n *
• min
:
∑ <,<> ∈? @AB
C ! C − ' C( E
s.t. * % ≤ G
* %
@AB
C #%
H #%
26
LIME
n
n
• vs
•
→ LIME
27
SHAP
n A Unified Approach to Interpreting Model Predictions,
NIPS'17 [Python SHAP]
•
• LIME
28
SHAP
n 1
n 2
29
SHAP
n SHAP LIME
• Adult One-Hot
n . “ ”
SHAP “ ”
30
SHAP
n SHAP = LIME
• ! "# = %& + %("# = %& + ∑* %*"*
#
3
n 1. + " = ! "#
•
31
SHAP
n SHAP = LIME
• ! "# = %& + %("# = %& + ∑* %*"*
#
3
n 2. "*
#
= 0 ⇒ %* = 0
•
32
SHAP
n SHAP = LIME
• ! "# = %& + %("# = %& + ∑* %*"*
#
3
n 3. + " = +, "# + +′
+,
# "# − +,
# "# ∖ 0 ≥ +, "# − +, "# ∖ 0 ⇒ %* +# ≥ %* +
• + +′ "*
#
+′ +
+# "*
#
+
33
SHAP
n SHAP = LIME
• ! "# = %& + %("# = %& + ∑* %*"*
#
SHAP
n 3
• %* = ∑+⊆-.
+ ! 01 + 12 !
0!
(4- 5 − 4-(5 ∖ 8))
• Shapley Value
34
SHAP
n LIME
• min
$
∑&∈( ) * +, * − . *
/
• ) * =
123
1 456678 & & (1 2|&|)
n
• Linear SHAP:
• Tree SHAP:
• Deep SHAP: DeepLIFT
35
Anchor
n Anchors: High-Precision Model-Agnostic Explanations,
AAAI'18 [Python Anchor]
•
36
Anchor
37
Anchor
n Anchor LIME
• Adult One-Hot
n
• Anchor +
38
Anchor
n “ +
•
39
Anchor
n !
• " ! ! " = 1
n %(⋅ |!) !
n ! =Anchor
• Anchor ! * " +
,- * ! 1. / 0. 1 ≥ +, ! " = 1
Anchor ! "
40
Anchor
n !" # $ 1& ' (& ) ≥ +
• 1 − -
Pr !" # $ 1& ' (& ) ≥ + ≥ 1 − -
41
Anchor
n ! =
!
• !
• max
%
&' ( [!(+)] s.t. Pr &' + ! 11 ( 21 3 ≥ 5 ≥ 1 − 7
n 1.
n 2.
• Pr &' + ! 11 ( 21 3 ≥ 5 1 − 7 !
42
n
n
• LIME, SHAP, Anchor
n
• influence
43
influence
n Understanding Black-box Predictions via Influence
Functions, ICML’17 [Python influence-release]
n ("′, %′)
"
44
influence
n ("′, %′)
"
n % = (("; *+), *+
*+ = argmin
2∈4
5
67(8,9)∈:
;(<; +)
*+=6> = argmin
2∈4
5
6∈: ?@A 6B6>
;(<; +)
n <> = ("′, %′) *+
• C+=6> − C+
*+=6> − *+ <>
45
<>
= ("′, %′)
influence
n !" = (%′, (′) *+
• ,+-." − ,+
*+-." − *+ !"
n
• !" = (%′, (′) ,+-."
•
n influence
• ,+-." − ,+
,+-." − ,+ ≈ −
1
2
345
-6
78(!"; ,+)
46
influence
n Data Poisoning
•
•
47
influence
n
•
48
This looks like that: deep learning for interpretable
image recognition, arxiv: 1806.10574.
n
n
• LIME
• SHAP
• Anchor
n
• influence
49
n
• Interpretable Predictions of Tree-based Ensembles
via Actionable Feature Tweaking, KDD’17
•
50
30 200
20 200
30 300
⭕
⭕
…
…
n
• Generating Visual Explanations, ECCV‘16
•
51
52
n
n Born Again Trees
n Making Tree Ensembles Interpretable: A Bayesian Model
Selection Approach, AISTATS'18 [Python defragTrees]
53
BATrees
n Born Again Trees
•
•
n
• !
!
•
54
n
• BATrees
• defragTrees RandomForest
55
56
n
n [Python+Tensorflow
saliency; DeepExplain]
• Striving for Simplicity: The All Convolutional Net
(GuidedBackprop)
• On Pixel-Wise Explanations for Non-Linear Classifier Decisions
by Layer-Wise Relevance Propagation (Epsilon-LRP)
• Axiomatic Attribution for Deep Networks (IntegratedGrad)
• SmoothGrad: Removing Noise by Adding Noise (SmoothGrad)
• Learning Important Features Through Propagating Activation
Differences (DeepLIFT)
57
n
•
58
DNN
n
•
59
DNN
n
•
60
DNN
n DNN
61
n
•
→
•
→
62
n ! = # $
n $
n [Simonyan et al., arXiv’14]
$%
&' (
&()
•
→ →
*+ ,
*,-
→
•
→ →
&' (
&()
→
63
n [Simonyan et al., arXiv’14]
!"
#$ %
#%&
n
• GuidedBP [Springenberg et al., arXiv’14]
back propagation
• LRP [Bach et al., PloS ONE’15]
• IntegratedGrad [Sundararajan et al., arXiv’17]
• SmoothGrad [Smilkov et al., arXiv’17]
• DeepLIFT [Shrikumar et al., ICML’17]
64
n
n
[Python+Tensorflow saliency; DeepExplain]
• Striving for Simplicity: The All Convolutional Net
(GuidedBackprop)
• On Pixel-Wise Explanations for Non-Linear Classifier Decisions
by Layer-Wise Relevance Propagation (Epsilon-LRP)
• Axiomatic Attribution for Deep Networks (IntegratedGrad)
• SmoothGrad: Removing Noise by Adding Noise (SmoothGrad)
• Learning Important Features Through Propagating Activation
Differences (DeepLIFT)
65
•
•
•
• AISTATS’18
• AAAI’17,18
66
n AISTATS’18
•
n AAAI’17,18
•
67
defragTrees
n Making Tree Ensembles Interpretable: A Bayesian Model
Selection Approach, AISTATS'18 [Python defragTrees]
•
•
n
69
when
Relationship ≠ Not-in-family, Wife
Capital Gain < 7370
when
Relationship ≠ Not-in-family
Capital Gain >= 7370
when
Relationship ≠ Not-in-family, Unmarried
Capital Gain < 5095
Capital Loss < 2114
when
Relationship = Not-in-family
Country ≠ China, Peru
Capital Gain < 5095
when
Relationship ≠ Not-in-family
Country ≠ China
Capital Gain < 5095
when
Relationship ≠ Not-in-family
Capital Gain >= 7370
…
…
n
n
•
70
y = XOR(x1 < 0.5, x2 < 0.5) + ✏
n
n
71
2017 The State of Data Science & Machine LearningYour Year on Kaggle: Most Memorable
Community Stats from 2016
n
n
72
R
n
•
•
n
73
defragTrees
n 1.
• !" #, % &)
n 2. !"( #, % ))
• !"( #, % )) ≈ !"(#, %|&) ) ≪ &
!"( #, % ))
n
• )
• Factorized
Asymptotic Bayesian (FAB) Inference
74
&
)
n
n
•
•
•
75
D
Synthetic 2 1000 1000
Spambase 57 1000 1000
MiniBooNE 50 5000 5000
Magic 11 5000 5000
Higgs 28 5000 5000
Energy 8 384 384
n
76
n
77
n
•
n
•
•
78
n AISTATS’18
•
n AAAI’17,18
•
79
n Enumerate Lasso Solu/ons for Feature Selec/on,
AAAI’17 [Python LassoVariants].
n Approximate and Exact Enumera/on of Rule Models,
AAAI'18.
→ NO!
n
•
82
n
• →
•
n
•
83
n
•
n
•
→
84
n
•
n
•
→
85
n
•
n
•
→
86
n
•
n
•
87
•
•
Lasso
Given: !", $" ∈ ℝ'×ℝ ) = 1, 2, … , .
Find: / ∈ ℝ' s.t. !"
0
/ ≈ $ () = 1, 2, … , .)
/
n
•
•
Lasso ℓ5
/∗ = argmin
=
1
2
>/ − $ @ + B / 5
• Lasso /∗ supp(/∗) = {) ∶ /"
∗
≠ 0}
Lasso
n
•
→ Lasso
n Lasso
•
→
n
Lasso
n ! ⊆ {$%, $', … , $)} Lasso
Lasso ! = min
3
%
'
45 − 7 ' + 9 5 % s.t. supp 5 ⊆ !
Lasso
! Lasso ! <
supp 5 = !
• 9
• 9
• $%, $', $= , $%, $', $> , $%, $>, $? , $%, $' , …
Lasso !
Lawler !-best
1. " #
2. $ ∈ #
" $ "& = " ∖ {$}
Lasso("′) #′
(#&, "′)
3.
4.
Lawler !-best
1. " #
2. $ ∈ &
' $ '( = ' ∖ {$}
Lasso('′) &′
(&(, '′)
3.
4.
& = 56, 57, 58
' = 56, 57, 59, 58, 5:
&6 = 56, 57, 58
'6 = 56, 57, 59, 58, 5:
Lawler !-best
1. " #
2. $ ∈ &
' $ '( = ' ∖ {$}
Lasso("′) #′
(#(, "′)
3.
4.
"5
( = 67, 68, 69, 6:
"7
( = 65, 68, 69, 6:
"8
(
= 65, 67, 68, 6:
"5 = 65, 67, 68, 69, 6:
#5 = 65, 67, 69
"5 = 65, 67, 68, 69, 6:
Lawler !-best
1. " #
2. $ ∈ &
" ' "( = " ∖ {'}
-.//0(2′) &′
(&(, 2′)
3.
4.
(#6
(= 78, 79, 7: , "6
()"6
( = 78, 7;, 79, 7:
"8
( = 76, 7;, 79, 7:
";
(
= 76, 78, 7;, 7:
"6 = 76, 78, 7;, 79, 7:
#6 = 76, 78, 79
"6 = 76, 78, 7;, 79, 7:
Lawler !-best
1. " #
2. $ ∈ &
" ' "( = " ∖ {'}
-.//0(2′) &′
(&(, 2′)
3.
4.
(#6
(= 78, 79, 7: , "6
()"6
( = 78, 7;, 79, 7:
"8
( = 76, 7;, 79, 7:
";
(
= 76, 78, 7;, 7:
"6 = 76, 78, 7;, 79, 7:
(#8
(= 76, 7;, 79 , "8
()
#6 = 76, 78, 79
"6 = 76, 78, 7;, 79, 7:
Lawler !-best
1. " #
2. $ ∈ &
" ' "( = " ∖ {'}
-.//0(2′) &′
(&(, 2′)
3.
4.
(#6
(= 78, 79, 7: , "6
()"6
( = 78, 7;, 79, 7:
"8
( = 76, 7;, 79, 7:
";
(
= 76, 78, 7;, 7:
"6 = 76, 78, 7;, 79, 7:
(#8
(= 76, 7;, 79 , "8
()
(#;
(= 76, 78, 7: , ";
()#6 = 76, 78, 79
"6 = 76, 78, 7;, 79, 7:
Lawler !-best
1. " #
2. $ ∈ #
" $ "& = " ∖ {$}
+,--.("′) #′
(#&, "′)
3.
4.
(#3
&= 45, 46, 47 , "3
&)
(#5
&= 43, 48, 46 , "5
&)
(#8
&= 43, 45, 47 , "8
&)
#5 = 45, 46, 47
"5 = 45, 48, 46, 47
#3 = 43, 45, 46
"3 = 43, 45, 48, 46, 47
Lawler !-best
1. " #
2. $ ∈ &
' $ '( = ' ∖ {$}
Lasso("′) #′
(#(, "′)
3.
4.
#5 = 65, 67, 68
"5 = 65, 67, 69, 68, 6:
(#7
(= 65, 69, 68 , "7
()
(#9
(= 65, 67, 6: , "9
()
#7 = 67, 68, 6:
"7 = 67, 69, 68, 6:
"8
( = 69, 68, 6:
":
(
= 67, 69, 6:
";
(
= 67, 69, 68
"7 = 67, 69, 68, 6:
Lasso % %
n
• %
• % Lasso
1.
n Thaliana gene expression data (Atwell et al. ’10):
• ! ∈ ℝ$%&%'( 2
• ) ∈ ℝ
• 134
2.
n 20 Newsgroups Data (Lang’95); ibm vs mac
• ! ∈ ℝ$$%&' tf-idf
• ( ∈ {ibm, mac} 2
• 1168
→
bios drive ibm
ide drive ibm
dos os, drive ibm
controller drive ibm
quadra, centris 040, clock mac
windows, bios, controller disk, drive ibm
bios, help, controller disk, drive ibm
centris, pc 610 mac
n
•
n Lasso
• Lawler !-best
n
•
•
•
•
•
• AISTATS’18
• AAAI’17,18
106
n
•
•
107
n
n Sanity Checks for Saliency Maps, NeurIPS’18.
•
• Sanity Check
108
“ ”
Guided-BP
n
•
•
•
n
• Please Stop Explaining Black Box Models for High-Stakes
Decisions, arXiv:1811.10154
109
n
•
•
n
•
110

More Related Content

What's hot

Anomaly detection 系の論文を一言でまとめた
Anomaly detection 系の論文を一言でまとめたAnomaly detection 系の論文を一言でまとめた
Anomaly detection 系の論文を一言でまとめたぱんいち すみもと
 
協力ゲーム理論でXAI (説明可能なAI) を目指すSHAP (Shapley Additive exPlanation)
協力ゲーム理論でXAI (説明可能なAI) を目指すSHAP (Shapley Additive exPlanation)協力ゲーム理論でXAI (説明可能なAI) を目指すSHAP (Shapley Additive exPlanation)
協力ゲーム理論でXAI (説明可能なAI) を目指すSHAP (Shapley Additive exPlanation)西岡 賢一郎
 
[DL輪読会]Understanding Black-box Predictions via Influence Functions
[DL輪読会]Understanding Black-box Predictions via Influence Functions [DL輪読会]Understanding Black-box Predictions via Influence Functions
[DL輪読会]Understanding Black-box Predictions via Influence Functions Deep Learning JP
 
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
 
Layer Normalization@NIPS+読み会・関西
Layer Normalization@NIPS+読み会・関西Layer Normalization@NIPS+読み会・関西
Layer Normalization@NIPS+読み会・関西Keigo Nishida
 
SSII2020TS: 機械学習モデルの判断根拠の説明​ 〜 Explainable AI 研究の近年の展開 〜​
SSII2020TS: 機械学習モデルの判断根拠の説明​ 〜 Explainable AI 研究の近年の展開 〜​SSII2020TS: 機械学習モデルの判断根拠の説明​ 〜 Explainable AI 研究の近年の展開 〜​
SSII2020TS: 機械学習モデルの判断根拠の説明​ 〜 Explainable AI 研究の近年の展開 〜​SSII
 
Triplet Loss 徹底解説
Triplet Loss 徹底解説Triplet Loss 徹底解説
Triplet Loss 徹底解説tancoro
 
機械学習による統計的実験計画(ベイズ最適化を中心に)
機械学習による統計的実験計画(ベイズ最適化を中心に)機械学習による統計的実験計画(ベイズ最適化を中心に)
機械学習による統計的実験計画(ベイズ最適化を中心に)Kota Matsui
 
【DL輪読会】High-Resolution Image Synthesis with Latent Diffusion Models
【DL輪読会】High-Resolution Image Synthesis with Latent Diffusion Models【DL輪読会】High-Resolution Image Synthesis with Latent Diffusion Models
【DL輪読会】High-Resolution Image Synthesis with Latent Diffusion ModelsDeep Learning JP
 
SSII2021 [OS2-01] 転移学習の基礎:異なるタスクの知識を利用するための機械学習の方法
SSII2021 [OS2-01] 転移学習の基礎:異なるタスクの知識を利用するための機械学習の方法SSII2021 [OS2-01] 転移学習の基礎:異なるタスクの知識を利用するための機械学習の方法
SSII2021 [OS2-01] 転移学習の基礎:異なるタスクの知識を利用するための機械学習の方法SSII
 
全力解説!Transformer
全力解説!Transformer全力解説!Transformer
全力解説!TransformerArithmer Inc.
 
【論文紹介】How Powerful are Graph Neural Networks?
【論文紹介】How Powerful are Graph Neural Networks?【論文紹介】How Powerful are Graph Neural Networks?
【論文紹介】How Powerful are Graph Neural Networks?Masanao Ochi
 
自己教師学習(Self-Supervised Learning)
自己教師学習(Self-Supervised Learning)自己教師学習(Self-Supervised Learning)
自己教師学習(Self-Supervised Learning)cvpaper. challenge
 
深層学習の数理
深層学習の数理深層学習の数理
深層学習の数理Taiji Suzuki
 
モデル高速化百選
モデル高速化百選モデル高速化百選
モデル高速化百選Yusuke Uchida
 
Optimizer入門&最新動向
Optimizer入門&最新動向Optimizer入門&最新動向
Optimizer入門&最新動向Motokawa Tetsuya
 
深層生成モデルと世界モデル
深層生成モデルと世界モデル深層生成モデルと世界モデル
深層生成モデルと世界モデルMasahiro Suzuki
 
“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
 
最近のKaggleに学ぶテーブルデータの特徴量エンジニアリング
最近のKaggleに学ぶテーブルデータの特徴量エンジニアリング最近のKaggleに学ぶテーブルデータの特徴量エンジニアリング
最近のKaggleに学ぶテーブルデータの特徴量エンジニアリングmlm_kansai
 

What's hot (20)

Anomaly detection 系の論文を一言でまとめた
Anomaly detection 系の論文を一言でまとめたAnomaly detection 系の論文を一言でまとめた
Anomaly detection 系の論文を一言でまとめた
 
協力ゲーム理論でXAI (説明可能なAI) を目指すSHAP (Shapley Additive exPlanation)
協力ゲーム理論でXAI (説明可能なAI) を目指すSHAP (Shapley Additive exPlanation)協力ゲーム理論でXAI (説明可能なAI) を目指すSHAP (Shapley Additive exPlanation)
協力ゲーム理論でXAI (説明可能なAI) を目指すSHAP (Shapley Additive exPlanation)
 
[DL輪読会]Understanding Black-box Predictions via Influence Functions
[DL輪読会]Understanding Black-box Predictions via Influence Functions [DL輪読会]Understanding Black-box Predictions via Influence Functions
[DL輪読会]Understanding Black-box Predictions via Influence Functions
 
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
 
Layer Normalization@NIPS+読み会・関西
Layer Normalization@NIPS+読み会・関西Layer Normalization@NIPS+読み会・関西
Layer Normalization@NIPS+読み会・関西
 
SSII2020TS: 機械学習モデルの判断根拠の説明​ 〜 Explainable AI 研究の近年の展開 〜​
SSII2020TS: 機械学習モデルの判断根拠の説明​ 〜 Explainable AI 研究の近年の展開 〜​SSII2020TS: 機械学習モデルの判断根拠の説明​ 〜 Explainable AI 研究の近年の展開 〜​
SSII2020TS: 機械学習モデルの判断根拠の説明​ 〜 Explainable AI 研究の近年の展開 〜​
 
Triplet Loss 徹底解説
Triplet Loss 徹底解説Triplet Loss 徹底解説
Triplet Loss 徹底解説
 
機械学習による統計的実験計画(ベイズ最適化を中心に)
機械学習による統計的実験計画(ベイズ最適化を中心に)機械学習による統計的実験計画(ベイズ最適化を中心に)
機械学習による統計的実験計画(ベイズ最適化を中心に)
 
【DL輪読会】High-Resolution Image Synthesis with Latent Diffusion Models
【DL輪読会】High-Resolution Image Synthesis with Latent Diffusion Models【DL輪読会】High-Resolution Image Synthesis with Latent Diffusion Models
【DL輪読会】High-Resolution Image Synthesis with Latent Diffusion Models
 
SSII2021 [OS2-01] 転移学習の基礎:異なるタスクの知識を利用するための機械学習の方法
SSII2021 [OS2-01] 転移学習の基礎:異なるタスクの知識を利用するための機械学習の方法SSII2021 [OS2-01] 転移学習の基礎:異なるタスクの知識を利用するための機械学習の方法
SSII2021 [OS2-01] 転移学習の基礎:異なるタスクの知識を利用するための機械学習の方法
 
全力解説!Transformer
全力解説!Transformer全力解説!Transformer
全力解説!Transformer
 
【論文紹介】How Powerful are Graph Neural Networks?
【論文紹介】How Powerful are Graph Neural Networks?【論文紹介】How Powerful are Graph Neural Networks?
【論文紹介】How Powerful are Graph Neural Networks?
 
自己教師学習(Self-Supervised Learning)
自己教師学習(Self-Supervised Learning)自己教師学習(Self-Supervised Learning)
自己教師学習(Self-Supervised Learning)
 
深層学習の数理
深層学習の数理深層学習の数理
深層学習の数理
 
モデル高速化百選
モデル高速化百選モデル高速化百選
モデル高速化百選
 
Optimizer入門&最新動向
Optimizer入門&最新動向Optimizer入門&最新動向
Optimizer入門&最新動向
 
深層生成モデルと世界モデル
深層生成モデルと世界モデル深層生成モデルと世界モデル
深層生成モデルと世界モデル
 
MICの解説
MICの解説MICの解説
MICの解説
 
“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
 
最近のKaggleに学ぶテーブルデータの特徴量エンジニアリング
最近のKaggleに学ぶテーブルデータの特徴量エンジニアリング最近のKaggleに学ぶテーブルデータの特徴量エンジニアリング
最近のKaggleに学ぶテーブルデータの特徴量エンジニアリング
 

Similar to 機械学習モデルの判断根拠の説明

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
 
Active Neural Localization
Active Neural LocalizationActive Neural Localization
Active Neural LocalizationHaruya Ishikawa
 
(論文サーベイ)主成分分析によるオンライン外れ値検出
(論文サーベイ)主成分分析によるオンライン外れ値検出(論文サーベイ)主成分分析によるオンライン外れ値検出
(論文サーベイ)主成分分析によるオンライン外れ値検出kazuya_n
 
Mcs011 solved assignment by divya singh
Mcs011 solved assignment by divya singhMcs011 solved assignment by divya singh
Mcs011 solved assignment by divya singhDIVYA SINGH
 
Basics in algorithms and data structure
Basics in algorithms and data structure Basics in algorithms and data structure
Basics in algorithms and data structure Eman magdy
 
(RubyConf 2020) How prime numbers keep the internet secure
(RubyConf 2020) How prime numbers keep the internet secure(RubyConf 2020) How prime numbers keep the internet secure
(RubyConf 2020) How prime numbers keep the internet secureSun-Li Beatteay
 
Elasticsearch at EyeEm
Elasticsearch at EyeEmElasticsearch at EyeEm
Elasticsearch at EyeEmLars Fronius
 
K02-salen: Systems Thinking in Action 2011
K02-salen: Systems Thinking in Action 2011K02-salen: Systems Thinking in Action 2011
K02-salen: Systems Thinking in Action 2011pegasuscomm
 
Oceans 2019 tutorial-geophysical-nav_7-updated
Oceans 2019 tutorial-geophysical-nav_7-updatedOceans 2019 tutorial-geophysical-nav_7-updated
Oceans 2019 tutorial-geophysical-nav_7-updatedFrancisco Curado-Teixeira
 
第5回NIPS読み会・関西発表資料
第5回NIPS読み会・関西発表資料第5回NIPS読み会・関西発表資料
第5回NIPS読み会・関西発表資料Kyoichiro Kobayashi
 
Paper Reading: Pessimistic Cardinality Estimation
Paper Reading: Pessimistic Cardinality EstimationPaper Reading: Pessimistic Cardinality Estimation
Paper Reading: Pessimistic Cardinality EstimationPingCAP
 
Moment-Based Estimation for Hierarchical Models in Apache Spark with Kyle Sch...
Moment-Based Estimation for Hierarchical Models in Apache Spark with Kyle Sch...Moment-Based Estimation for Hierarchical Models in Apache Spark with Kyle Sch...
Moment-Based Estimation for Hierarchical Models in Apache Spark with Kyle Sch...Databricks
 
Duplicates everywhere (Kiev)
Duplicates everywhere (Kiev)Duplicates everywhere (Kiev)
Duplicates everywhere (Kiev)Alexey Grigorev
 
Rob Sullivan at Heroku's Waza 2013: Your Database -- A Story of Indifference
Rob Sullivan at Heroku's Waza 2013: Your Database -- A Story of IndifferenceRob Sullivan at Heroku's Waza 2013: Your Database -- A Story of Indifference
Rob Sullivan at Heroku's Waza 2013: Your Database -- A Story of IndifferenceHeroku
 
A3 sec -_regular_expressions
A3 sec -_regular_expressionsA3 sec -_regular_expressions
A3 sec -_regular_expressionsa3sec
 
(KO) 온라인 뉴스 댓글 플랫폼을 흐리는 어뷰저 분석기 / (EN) Online ...
(KO) 온라인 뉴스 댓글 플랫폼을 흐리는 어뷰저 분석기 / (EN) Online ...(KO) 온라인 뉴스 댓글 플랫폼을 흐리는 어뷰저 분석기 / (EN) Online ...
(KO) 온라인 뉴스 댓글 플랫폼을 흐리는 어뷰저 분석기 / (EN) Online ...Ji Hyung Moon
 
Datamining R 2nd
Datamining R 2ndDatamining R 2nd
Datamining R 2ndsesejun
 
Parallel Computing in R
Parallel Computing in RParallel Computing in R
Parallel Computing in Rmickey24
 
Locality sensitive hashing
Locality sensitive hashingLocality sensitive hashing
Locality sensitive hashingSEMINARGROOT
 

Similar to 機械学習モデルの判断根拠の説明 (20)

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...
 
Active Neural Localization
Active Neural LocalizationActive Neural Localization
Active Neural Localization
 
機械学習と自動微分
機械学習と自動微分機械学習と自動微分
機械学習と自動微分
 
(論文サーベイ)主成分分析によるオンライン外れ値検出
(論文サーベイ)主成分分析によるオンライン外れ値検出(論文サーベイ)主成分分析によるオンライン外れ値検出
(論文サーベイ)主成分分析によるオンライン外れ値検出
 
Mcs011 solved assignment by divya singh
Mcs011 solved assignment by divya singhMcs011 solved assignment by divya singh
Mcs011 solved assignment by divya singh
 
Basics in algorithms and data structure
Basics in algorithms and data structure Basics in algorithms and data structure
Basics in algorithms and data structure
 
(RubyConf 2020) How prime numbers keep the internet secure
(RubyConf 2020) How prime numbers keep the internet secure(RubyConf 2020) How prime numbers keep the internet secure
(RubyConf 2020) How prime numbers keep the internet secure
 
Elasticsearch at EyeEm
Elasticsearch at EyeEmElasticsearch at EyeEm
Elasticsearch at EyeEm
 
K02-salen: Systems Thinking in Action 2011
K02-salen: Systems Thinking in Action 2011K02-salen: Systems Thinking in Action 2011
K02-salen: Systems Thinking in Action 2011
 
Oceans 2019 tutorial-geophysical-nav_7-updated
Oceans 2019 tutorial-geophysical-nav_7-updatedOceans 2019 tutorial-geophysical-nav_7-updated
Oceans 2019 tutorial-geophysical-nav_7-updated
 
第5回NIPS読み会・関西発表資料
第5回NIPS読み会・関西発表資料第5回NIPS読み会・関西発表資料
第5回NIPS読み会・関西発表資料
 
Paper Reading: Pessimistic Cardinality Estimation
Paper Reading: Pessimistic Cardinality EstimationPaper Reading: Pessimistic Cardinality Estimation
Paper Reading: Pessimistic Cardinality Estimation
 
Moment-Based Estimation for Hierarchical Models in Apache Spark with Kyle Sch...
Moment-Based Estimation for Hierarchical Models in Apache Spark with Kyle Sch...Moment-Based Estimation for Hierarchical Models in Apache Spark with Kyle Sch...
Moment-Based Estimation for Hierarchical Models in Apache Spark with Kyle Sch...
 
Duplicates everywhere (Kiev)
Duplicates everywhere (Kiev)Duplicates everywhere (Kiev)
Duplicates everywhere (Kiev)
 
Rob Sullivan at Heroku's Waza 2013: Your Database -- A Story of Indifference
Rob Sullivan at Heroku's Waza 2013: Your Database -- A Story of IndifferenceRob Sullivan at Heroku's Waza 2013: Your Database -- A Story of Indifference
Rob Sullivan at Heroku's Waza 2013: Your Database -- A Story of Indifference
 
A3 sec -_regular_expressions
A3 sec -_regular_expressionsA3 sec -_regular_expressions
A3 sec -_regular_expressions
 
(KO) 온라인 뉴스 댓글 플랫폼을 흐리는 어뷰저 분석기 / (EN) Online ...
(KO) 온라인 뉴스 댓글 플랫폼을 흐리는 어뷰저 분석기 / (EN) Online ...(KO) 온라인 뉴스 댓글 플랫폼을 흐리는 어뷰저 분석기 / (EN) Online ...
(KO) 온라인 뉴스 댓글 플랫폼을 흐리는 어뷰저 분석기 / (EN) Online ...
 
Datamining R 2nd
Datamining R 2ndDatamining R 2nd
Datamining R 2nd
 
Parallel Computing in R
Parallel Computing in RParallel Computing in R
Parallel Computing in R
 
Locality sensitive hashing
Locality sensitive hashingLocality sensitive hashing
Locality sensitive hashing
 

More from Satoshi Hara

Explanation in Machine Learning and Its Reliability
Explanation in Machine Learning and Its ReliabilityExplanation in Machine Learning and Its Reliability
Explanation in Machine Learning and Its ReliabilitySatoshi Hara
 
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
 
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 SolutionsSatoshi Hara
 
Maximally Invariant Data Perturbation as Explanation
Maximally Invariant Data Perturbation as ExplanationMaximally Invariant Data Perturbation as Explanation
Maximally Invariant Data Perturbation as ExplanationSatoshi 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 AutoencodersSatoshi 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 (8)

Explanation in Machine Learning and Its Reliability
Explanation in Machine Learning and Its ReliabilityExplanation in Machine Learning and Its Reliability
Explanation in Machine Learning and Its Reliability
 
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
 
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
 
Maximally Invariant Data Perturbation as Explanation
Maximally Invariant Data Perturbation as ExplanationMaximally Invariant Data Perturbation as Explanation
Maximally Invariant Data Perturbation as Explanation
 
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

Long journey of Ruby standard library at RubyConf AU 2024
Long journey of Ruby standard library at RubyConf AU 2024Long journey of Ruby standard library at RubyConf AU 2024
Long journey of Ruby standard library at RubyConf AU 2024Hiroshi SHIBATA
 
Microsoft 365 Copilot: How to boost your productivity with AI – Part two: Dat...
Microsoft 365 Copilot: How to boost your productivity with AI – Part two: Dat...Microsoft 365 Copilot: How to boost your productivity with AI – Part two: Dat...
Microsoft 365 Copilot: How to boost your productivity with AI – Part two: Dat...Nikki Chapple
 
MuleSoft Online Meetup Group - B2B Crash Course: Release SparkNotes
MuleSoft Online Meetup Group - B2B Crash Course: Release SparkNotesMuleSoft Online Meetup Group - B2B Crash Course: Release SparkNotes
MuleSoft Online Meetup Group - B2B Crash Course: Release SparkNotesManik S Magar
 
So einfach geht modernes Roaming fuer Notes und Nomad.pdf
So einfach geht modernes Roaming fuer Notes und Nomad.pdfSo einfach geht modernes Roaming fuer Notes und Nomad.pdf
So einfach geht modernes Roaming fuer Notes und Nomad.pdfpanagenda
 
Abdul Kader Baba- Managing Cybersecurity Risks and Compliance Requirements i...
Abdul Kader Baba- Managing Cybersecurity Risks  and Compliance Requirements i...Abdul Kader Baba- Managing Cybersecurity Risks  and Compliance Requirements i...
Abdul Kader Baba- Managing Cybersecurity Risks and Compliance Requirements i...itnewsafrica
 
UiPath Community: Communication Mining from Zero to Hero
UiPath Community: Communication Mining from Zero to HeroUiPath Community: Communication Mining from Zero to Hero
UiPath Community: Communication Mining from Zero to HeroUiPathCommunity
 
React JS; all concepts. Contains React Features, JSX, functional & Class comp...
React JS; all concepts. Contains React Features, JSX, functional & Class comp...React JS; all concepts. Contains React Features, JSX, functional & Class comp...
React JS; all concepts. Contains React Features, JSX, functional & Class comp...Karmanjay Verma
 
Email Marketing Automation for Bonterra Impact Management (fka Social Solutio...
Email Marketing Automation for Bonterra Impact Management (fka Social Solutio...Email Marketing Automation for Bonterra Impact Management (fka Social Solutio...
Email Marketing Automation for Bonterra Impact Management (fka Social Solutio...Jeffrey Haguewood
 
Digital Tools & AI in Career Development
Digital Tools & AI in Career DevelopmentDigital Tools & AI in Career Development
Digital Tools & AI in Career DevelopmentMahmoud Rabie
 
Arizona Broadband Policy Past, Present, and Future Presentation 3/25/24
Arizona Broadband Policy Past, Present, and Future Presentation 3/25/24Arizona Broadband Policy Past, Present, and Future Presentation 3/25/24
Arizona Broadband Policy Past, Present, and Future Presentation 3/25/24Mark Goldstein
 
QMMS Lesson 2 - Using MS Excel Formula.pdf
QMMS Lesson 2 - Using MS Excel Formula.pdfQMMS Lesson 2 - Using MS Excel Formula.pdf
QMMS Lesson 2 - Using MS Excel Formula.pdfROWELL MARQUINA
 
Landscape Catalogue 2024 Australia-1.pdf
Landscape Catalogue 2024 Australia-1.pdfLandscape Catalogue 2024 Australia-1.pdf
Landscape Catalogue 2024 Australia-1.pdfAarwolf Industries LLC
 
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...BookNet Canada
 
Top 10 Hubspot Development Companies in 2024
Top 10 Hubspot Development Companies in 2024Top 10 Hubspot Development Companies in 2024
Top 10 Hubspot Development Companies in 2024TopCSSGallery
 
2024 April Patch Tuesday
2024 April Patch Tuesday2024 April Patch Tuesday
2024 April Patch TuesdayIvanti
 
React Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App FrameworkReact Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App FrameworkPixlogix Infotech
 
JET Technology Labs White Paper for Virtualized Security and Encryption Techn...
JET Technology Labs White Paper for Virtualized Security and Encryption Techn...JET Technology Labs White Paper for Virtualized Security and Encryption Techn...
JET Technology Labs White Paper for Virtualized Security and Encryption Techn...amber724300
 
WomenInAutomation2024: AI and Automation for eveyone
WomenInAutomation2024: AI and Automation for eveyoneWomenInAutomation2024: AI and Automation for eveyone
WomenInAutomation2024: AI and Automation for eveyoneUiPathCommunity
 
Glenn Lazarus- Why Your Observability Strategy Needs Security Observability
Glenn Lazarus- Why Your Observability Strategy Needs Security ObservabilityGlenn Lazarus- Why Your Observability Strategy Needs Security Observability
Glenn Lazarus- Why Your Observability Strategy Needs Security Observabilityitnewsafrica
 
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...Wes McKinney
 

Recently uploaded (20)

Long journey of Ruby standard library at RubyConf AU 2024
Long journey of Ruby standard library at RubyConf AU 2024Long journey of Ruby standard library at RubyConf AU 2024
Long journey of Ruby standard library at RubyConf AU 2024
 
Microsoft 365 Copilot: How to boost your productivity with AI – Part two: Dat...
Microsoft 365 Copilot: How to boost your productivity with AI – Part two: Dat...Microsoft 365 Copilot: How to boost your productivity with AI – Part two: Dat...
Microsoft 365 Copilot: How to boost your productivity with AI – Part two: Dat...
 
MuleSoft Online Meetup Group - B2B Crash Course: Release SparkNotes
MuleSoft Online Meetup Group - B2B Crash Course: Release SparkNotesMuleSoft Online Meetup Group - B2B Crash Course: Release SparkNotes
MuleSoft Online Meetup Group - B2B Crash Course: Release SparkNotes
 
So einfach geht modernes Roaming fuer Notes und Nomad.pdf
So einfach geht modernes Roaming fuer Notes und Nomad.pdfSo einfach geht modernes Roaming fuer Notes und Nomad.pdf
So einfach geht modernes Roaming fuer Notes und Nomad.pdf
 
Abdul Kader Baba- Managing Cybersecurity Risks and Compliance Requirements i...
Abdul Kader Baba- Managing Cybersecurity Risks  and Compliance Requirements i...Abdul Kader Baba- Managing Cybersecurity Risks  and Compliance Requirements i...
Abdul Kader Baba- Managing Cybersecurity Risks and Compliance Requirements i...
 
UiPath Community: Communication Mining from Zero to Hero
UiPath Community: Communication Mining from Zero to HeroUiPath Community: Communication Mining from Zero to Hero
UiPath Community: Communication Mining from Zero to Hero
 
React JS; all concepts. Contains React Features, JSX, functional & Class comp...
React JS; all concepts. Contains React Features, JSX, functional & Class comp...React JS; all concepts. Contains React Features, JSX, functional & Class comp...
React JS; all concepts. Contains React Features, JSX, functional & Class comp...
 
Email Marketing Automation for Bonterra Impact Management (fka Social Solutio...
Email Marketing Automation for Bonterra Impact Management (fka Social Solutio...Email Marketing Automation for Bonterra Impact Management (fka Social Solutio...
Email Marketing Automation for Bonterra Impact Management (fka Social Solutio...
 
Digital Tools & AI in Career Development
Digital Tools & AI in Career DevelopmentDigital Tools & AI in Career Development
Digital Tools & AI in Career Development
 
Arizona Broadband Policy Past, Present, and Future Presentation 3/25/24
Arizona Broadband Policy Past, Present, and Future Presentation 3/25/24Arizona Broadband Policy Past, Present, and Future Presentation 3/25/24
Arizona Broadband Policy Past, Present, and Future Presentation 3/25/24
 
QMMS Lesson 2 - Using MS Excel Formula.pdf
QMMS Lesson 2 - Using MS Excel Formula.pdfQMMS Lesson 2 - Using MS Excel Formula.pdf
QMMS Lesson 2 - Using MS Excel Formula.pdf
 
Landscape Catalogue 2024 Australia-1.pdf
Landscape Catalogue 2024 Australia-1.pdfLandscape Catalogue 2024 Australia-1.pdf
Landscape Catalogue 2024 Australia-1.pdf
 
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
Transcript: New from BookNet Canada for 2024: BNC SalesData and LibraryData -...
 
Top 10 Hubspot Development Companies in 2024
Top 10 Hubspot Development Companies in 2024Top 10 Hubspot Development Companies in 2024
Top 10 Hubspot Development Companies in 2024
 
2024 April Patch Tuesday
2024 April Patch Tuesday2024 April Patch Tuesday
2024 April Patch Tuesday
 
React Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App FrameworkReact Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App Framework
 
JET Technology Labs White Paper for Virtualized Security and Encryption Techn...
JET Technology Labs White Paper for Virtualized Security and Encryption Techn...JET Technology Labs White Paper for Virtualized Security and Encryption Techn...
JET Technology Labs White Paper for Virtualized Security and Encryption Techn...
 
WomenInAutomation2024: AI and Automation for eveyone
WomenInAutomation2024: AI and Automation for eveyoneWomenInAutomation2024: AI and Automation for eveyone
WomenInAutomation2024: AI and Automation for eveyone
 
Glenn Lazarus- Why Your Observability Strategy Needs Security Observability
Glenn Lazarus- Why Your Observability Strategy Needs Security ObservabilityGlenn Lazarus- Why Your Observability Strategy Needs Security Observability
Glenn Lazarus- Why Your Observability Strategy Needs Security Observability
 
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
 

機械学習モデルの判断根拠の説明

  • 2. n • ~2013.3, PhD@ , • 2013.4~2016.3, @IBM • 2016.4~2017.8, @ ERATO, NII • 2017.9~, @ , n • ECML’11 AISTATS’15,17 • AISTATS’18 AAAI’17,18 ongoing 2
  • 8. 8
  • 9. EU GDPR n GDPR-22 1. The data subject shall have the right not to be subject to a decision based solely on automated processing, including profiling, which produces legal effects concerning him or her or similarly significantly affects him or her. 2. Paragraph 1 shall not apply if the decision: is necessary for entering into, or performance of, a contract between the data subject and a data controller; is authorised by Union or Member State law to which the controller is subject and which also lays down suitable measures to safeguard the data subject’s rights and freedoms and legitimate interests; or is based on the data subject’s explicit consent. 3. In the cases referred to in points (a) and (c) of paragraph 2, the data controller shall implement suitable measures to safeguard the data subject’s rights and freedoms and legitimate interests, at least the right to obtain human intervention on the part of the controller, to express his or her point of view and to contest the decision. 4. Decisions referred to in paragraph 2 shall not be based on special categories of personal data referred to in Article 9(2)1), unless point (a) or (g) of Article 9(2) applies and suitable measures to safeguard the data subject’s rights and freedoms and legitimate interests are in place. 9
  • 10. n 2016 • ICML, NIPS n • , , Vol.33, No.3, pages 366--369, 2018. • , Qiita 10
  • 11. n AI 11 Peeking inside the black-box: A survey on Explainable Artificial Intelligence (XAI) https://ieeexplore.ieee.org/document/8466590/ 7 SCOPUS, IEEExplore, ACM Digital Library, Google Scholar, Citeseer Library, ScienceDirect, arXiv “intelligible”, “interpretable”, “transparency”, “black box”, “understandable”, “comprehensible”, “explainable” AI “Artificial Intelligence”, “Intelligent system”, “Machine learning”, “deep learning”, “classifier” , “decision tree”
  • 13. n • n • 1. • 2. • 3. • 4. • … 13
  • 16. n n Why Should I Trust You?: Explaining the Predictions of Any Classifier, KDD'16 [Python LIME; R LIME] n A Unified Approach to Interpreting Model Predictions, NIPS'17 [Python SHAP] n Anchors: High-Precision Model-Agnostic Explanations, AAAI'18 [Python Anchor] n Understanding Black-box Predictions via Influence Functions, ICML’17 [Python influence-release] 16
  • 17. n n Born Again Trees n Making Tree Ensembles Interpretable: A Bayesian Model Selection Approach, AISTATS'18 [Python defragTrees] 17
  • 18. n n [Python+Tensorflow saliency; DeepExplain] • Striving for Simplicity: The All Convolutional Net (GuidedBackprop) • On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation (Epsilon-LRP) • Axiomatic Attribution for Deep Networks (IntegratedGrad) • SmoothGrad: Removing Noise by Adding Noise (SmoothGrad) • Learning Important Features Through Propagating Activation Differences (DeepLIFT) 18
  • 19. 19
  • 20. n n Why Should I Trust You?: Explaining the Predictions of Any Classifier, KDD'16 [Python LIME; R LIME] n A Unified Approach to Interpreting Model Predictions, NIPS'17 [Python SHAP] n Anchors: High-Precision Model-Agnostic Explanations, AAAI'18 [Python Anchor] n Understanding Black-box Predictions via Influence Functions, ICML’17 [Python influence-release] 20
  • 21. n n • LIME, SHAP, Anchor n • influence 21
  • 22. LIME n Why Should I Trust You?: Explaining the Predictions of Any Classifier, KDD'16 [Python LIME; R LIME] • • 22
  • 24. LIME n 2 • One-Hot • n LIME • Adult One-Hot 24
  • 25. LIME n ! !′ • LIME !′ !# ∈ {0, 1}* + !, # + n -(!) !0 • - ! ≈ 2 !# ≔ 40 + 46!# for !# ∈ NeighborOf(!0 # ) • 4 4, 25
  • 26. LIME n !(#) #% • ! # ≈ ' #( ≔ *% + *,#( for #( ∈ NeighborOf(#% ( ) n * • min : ∑ <,<> ∈? @AB C ! C − ' C( E s.t. * % ≤ G * % @AB C #% H #% 26
  • 28. SHAP n A Unified Approach to Interpreting Model Predictions, NIPS'17 [Python SHAP] • • LIME 28
  • 30. SHAP n SHAP LIME • Adult One-Hot n . “ ” SHAP “ ” 30
  • 31. SHAP n SHAP = LIME • ! "# = %& + %("# = %& + ∑* %*"* # 3 n 1. + " = ! "# • 31
  • 32. SHAP n SHAP = LIME • ! "# = %& + %("# = %& + ∑* %*"* # 3 n 2. "* # = 0 ⇒ %* = 0 • 32
  • 33. SHAP n SHAP = LIME • ! "# = %& + %("# = %& + ∑* %*"* # 3 n 3. + " = +, "# + +′ +, # "# − +, # "# ∖ 0 ≥ +, "# − +, "# ∖ 0 ⇒ %* +# ≥ %* + • + +′ "* # +′ + +# "* # + 33
  • 34. SHAP n SHAP = LIME • ! "# = %& + %("# = %& + ∑* %*"* # SHAP n 3 • %* = ∑+⊆-. + ! 01 + 12 ! 0! (4- 5 − 4-(5 ∖ 8)) • Shapley Value 34
  • 35. SHAP n LIME • min $ ∑&∈( ) * +, * − . * / • ) * = 123 1 456678 & & (1 2|&|) n • Linear SHAP: • Tree SHAP: • Deep SHAP: DeepLIFT 35
  • 36. Anchor n Anchors: High-Precision Model-Agnostic Explanations, AAAI'18 [Python Anchor] • 36
  • 38. Anchor n Anchor LIME • Adult One-Hot n • Anchor + 38
  • 40. Anchor n ! • " ! ! " = 1 n %(⋅ |!) ! n ! =Anchor • Anchor ! * " + ,- * ! 1. / 0. 1 ≥ +, ! " = 1 Anchor ! " 40
  • 41. Anchor n !" # $ 1& ' (& ) ≥ + • 1 − - Pr !" # $ 1& ' (& ) ≥ + ≥ 1 − - 41
  • 42. Anchor n ! = ! • ! • max % &' ( [!(+)] s.t. Pr &' + ! 11 ( 21 3 ≥ 5 ≥ 1 − 7 n 1. n 2. • Pr &' + ! 11 ( 21 3 ≥ 5 1 − 7 ! 42
  • 43. n n • LIME, SHAP, Anchor n • influence 43
  • 44. influence n Understanding Black-box Predictions via Influence Functions, ICML’17 [Python influence-release] n ("′, %′) " 44
  • 45. influence n ("′, %′) " n % = (("; *+), *+ *+ = argmin 2∈4 5 67(8,9)∈: ;(<; +) *+=6> = argmin 2∈4 5 6∈: ?@A 6B6> ;(<; +) n <> = ("′, %′) *+ • C+=6> − C+ *+=6> − *+ <> 45 <> = ("′, %′)
  • 46. influence n !" = (%′, (′) *+ • ,+-." − ,+ *+-." − *+ !" n • !" = (%′, (′) ,+-." • n influence • ,+-." − ,+ ,+-." − ,+ ≈ − 1 2 345 -6 78(!"; ,+) 46
  • 48. influence n • 48 This looks like that: deep learning for interpretable image recognition, arxiv: 1806.10574.
  • 49. n n • LIME • SHAP • Anchor n • influence 49
  • 50. n • Interpretable Predictions of Tree-based Ensembles via Actionable Feature Tweaking, KDD’17 • 50 30 200 20 200 30 300 ⭕ ⭕ … …
  • 51. n • Generating Visual Explanations, ECCV‘16 • 51
  • 52. 52
  • 53. n n Born Again Trees n Making Tree Ensembles Interpretable: A Bayesian Model Selection Approach, AISTATS'18 [Python defragTrees] 53
  • 54. BATrees n Born Again Trees • • n • ! ! • 54
  • 55. n • BATrees • defragTrees RandomForest 55
  • 56. 56
  • 57. n n [Python+Tensorflow saliency; DeepExplain] • Striving for Simplicity: The All Convolutional Net (GuidedBackprop) • On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation (Epsilon-LRP) • Axiomatic Attribution for Deep Networks (IntegratedGrad) • SmoothGrad: Removing Noise by Adding Noise (SmoothGrad) • Learning Important Features Through Propagating Activation Differences (DeepLIFT) 57
  • 63. n ! = # $ n $ n [Simonyan et al., arXiv’14] $% &' ( &() • → → *+ , *,- → • → → &' ( &() → 63
  • 64. n [Simonyan et al., arXiv’14] !" #$ % #%& n • GuidedBP [Springenberg et al., arXiv’14] back propagation • LRP [Bach et al., PloS ONE’15] • IntegratedGrad [Sundararajan et al., arXiv’17] • SmoothGrad [Smilkov et al., arXiv’17] • DeepLIFT [Shrikumar et al., ICML’17] 64
  • 65. n n [Python+Tensorflow saliency; DeepExplain] • Striving for Simplicity: The All Convolutional Net (GuidedBackprop) • On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation (Epsilon-LRP) • Axiomatic Attribution for Deep Networks (IntegratedGrad) • SmoothGrad: Removing Noise by Adding Noise (SmoothGrad) • Learning Important Features Through Propagating Activation Differences (DeepLIFT) 65
  • 68.
  • 69. defragTrees n Making Tree Ensembles Interpretable: A Bayesian Model Selection Approach, AISTATS'18 [Python defragTrees] • • n 69 when Relationship ≠ Not-in-family, Wife Capital Gain < 7370 when Relationship ≠ Not-in-family Capital Gain >= 7370 when Relationship ≠ Not-in-family, Unmarried Capital Gain < 5095 Capital Loss < 2114 when Relationship = Not-in-family Country ≠ China, Peru Capital Gain < 5095 when Relationship ≠ Not-in-family Country ≠ China Capital Gain < 5095 when Relationship ≠ Not-in-family Capital Gain >= 7370 … …
  • 70. n n • 70 y = XOR(x1 < 0.5, x2 < 0.5) + ✏
  • 71. n n 71 2017 The State of Data Science & Machine LearningYour Year on Kaggle: Most Memorable Community Stats from 2016
  • 74. defragTrees n 1. • !" #, % &) n 2. !"( #, % )) • !"( #, % )) ≈ !"(#, %|&) ) ≪ & !"( #, % )) n • ) • Factorized Asymptotic Bayesian (FAB) Inference 74 & )
  • 75. n n • • • 75 D Synthetic 2 1000 1000 Spambase 57 1000 1000 MiniBooNE 50 5000 5000 Magic 11 5000 5000 Higgs 28 5000 5000 Energy 8 384 384
  • 76. n 76
  • 77. n 77
  • 80.
  • 81. n Enumerate Lasso Solu/ons for Feature Selec/on, AAAI’17 [Python LassoVariants]. n Approximate and Exact Enumera/on of Rule Models, AAAI'18.
  • 89. Lasso Given: !", $" ∈ ℝ'×ℝ ) = 1, 2, … , . Find: / ∈ ℝ' s.t. !" 0 / ≈ $ () = 1, 2, … , .) / n • • Lasso ℓ5 /∗ = argmin = 1 2 >/ − $ @ + B / 5 • Lasso /∗ supp(/∗) = {) ∶ /" ∗ ≠ 0}
  • 91.
  • 92. Lasso n ! ⊆ {$%, $', … , $)} Lasso Lasso ! = min 3 % ' 45 − 7 ' + 9 5 % s.t. supp 5 ⊆ ! Lasso ! Lasso ! < supp 5 = ! • 9 • 9 • $%, $', $= , $%, $', $> , $%, $>, $? , $%, $' , … Lasso !
  • 93. Lawler !-best 1. " # 2. $ ∈ # " $ "& = " ∖ {$} Lasso("′) #′ (#&, "′) 3. 4.
  • 94. Lawler !-best 1. " # 2. $ ∈ & ' $ '( = ' ∖ {$} Lasso('′) &′ (&(, '′) 3. 4. & = 56, 57, 58 ' = 56, 57, 59, 58, 5: &6 = 56, 57, 58 '6 = 56, 57, 59, 58, 5:
  • 95. Lawler !-best 1. " # 2. $ ∈ & ' $ '( = ' ∖ {$} Lasso("′) #′ (#(, "′) 3. 4. "5 ( = 67, 68, 69, 6: "7 ( = 65, 68, 69, 6: "8 ( = 65, 67, 68, 6: "5 = 65, 67, 68, 69, 6: #5 = 65, 67, 69 "5 = 65, 67, 68, 69, 6:
  • 96. Lawler !-best 1. " # 2. $ ∈ & " ' "( = " ∖ {'} -.//0(2′) &′ (&(, 2′) 3. 4. (#6 (= 78, 79, 7: , "6 ()"6 ( = 78, 7;, 79, 7: "8 ( = 76, 7;, 79, 7: "; ( = 76, 78, 7;, 7: "6 = 76, 78, 7;, 79, 7: #6 = 76, 78, 79 "6 = 76, 78, 7;, 79, 7:
  • 97. Lawler !-best 1. " # 2. $ ∈ & " ' "( = " ∖ {'} -.//0(2′) &′ (&(, 2′) 3. 4. (#6 (= 78, 79, 7: , "6 ()"6 ( = 78, 7;, 79, 7: "8 ( = 76, 7;, 79, 7: "; ( = 76, 78, 7;, 7: "6 = 76, 78, 7;, 79, 7: (#8 (= 76, 7;, 79 , "8 () #6 = 76, 78, 79 "6 = 76, 78, 7;, 79, 7:
  • 98. Lawler !-best 1. " # 2. $ ∈ & " ' "( = " ∖ {'} -.//0(2′) &′ (&(, 2′) 3. 4. (#6 (= 78, 79, 7: , "6 ()"6 ( = 78, 7;, 79, 7: "8 ( = 76, 7;, 79, 7: "; ( = 76, 78, 7;, 7: "6 = 76, 78, 7;, 79, 7: (#8 (= 76, 7;, 79 , "8 () (#; (= 76, 78, 7: , "; ()#6 = 76, 78, 79 "6 = 76, 78, 7;, 79, 7:
  • 99. Lawler !-best 1. " # 2. $ ∈ # " $ "& = " ∖ {$} +,--.("′) #′ (#&, "′) 3. 4. (#3 &= 45, 46, 47 , "3 &) (#5 &= 43, 48, 46 , "5 &) (#8 &= 43, 45, 47 , "8 &) #5 = 45, 46, 47 "5 = 45, 48, 46, 47 #3 = 43, 45, 46 "3 = 43, 45, 48, 46, 47
  • 100. Lawler !-best 1. " # 2. $ ∈ & ' $ '( = ' ∖ {$} Lasso("′) #′ (#(, "′) 3. 4. #5 = 65, 67, 68 "5 = 65, 67, 69, 68, 6: (#7 (= 65, 69, 68 , "7 () (#9 (= 65, 67, 6: , "9 () #7 = 67, 68, 6: "7 = 67, 69, 68, 6: "8 ( = 69, 68, 6: ": ( = 67, 69, 6: "; ( = 67, 69, 68 "7 = 67, 69, 68, 6:
  • 101. Lasso % % n • % • % Lasso
  • 102.
  • 103. 1. n Thaliana gene expression data (Atwell et al. ’10): • ! ∈ ℝ$%&%'( 2 • ) ∈ ℝ • 134
  • 104. 2. n 20 Newsgroups Data (Lang’95); ibm vs mac • ! ∈ ℝ$$%&' tf-idf • ( ∈ {ibm, mac} 2 • 1168 → bios drive ibm ide drive ibm dos os, drive ibm controller drive ibm quadra, centris 040, clock mac windows, bios, controller disk, drive ibm bios, help, controller disk, drive ibm centris, pc 610 mac
  • 105. n • n Lasso • Lawler !-best n • •
  • 108. n n Sanity Checks for Saliency Maps, NeurIPS’18. • • Sanity Check 108 “ ” Guided-BP
  • 109. n • • • n • Please Stop Explaining Black Box Models for High-Stakes Decisions, arXiv:1811.10154 109