[Paper reading] L-SHAPLEY AND C-SHAPLEY: EFFICIENT MODEL INTERPRETATION FOR STRUCTURED DATA

L-SHAPLEY and C-SHAPLEY: Efﬁcient
INTERPRETATION FOR STRUCTURED DATA
Kyoto University
Daiki Tanaka

Background
• Although many black-box ML models, such as
RandomForest, NN or kernel-methods, can produce highly
accurate prediction, but such predictions are lack of
interpretability.
1. Luck of interpretation is a crucial issue when black-box
models are applied in areas such as medicine, ﬁnancial
markets, and criminal justice.
2. To be able to know the model’s reasoning is the good
way to improve the model.

Background : There are kinds of approaches for
interpreting models.
• Model-speciﬁc interpretation or Model-agnostic interpretation
• Model-speciﬁc : making some assumptions to models. (e.g. methods based on
attention weights, or gradient based method like smooth-grad, grad-CAM, …)
• Model-agnostic : making no assumption to models. (e.g. LIME, or Shapley
value)
• Model-level interpretation or Instance-wise interpretation
• Instance-wise : yielding feature importances for each input instance (e.g.
saliency map).
• Model-level : yielding feature importances for the whole model. (e.g. weights
of Logistic Regression, or decision rules of decision tree)
This study focuses on Model-Agnostic & Instance-wise interpretation.

Problem Setting : Model-Agnostic & Instance-wise
interpretation
• Input :
• An Instance
• A predictive model
• Output :
• A vector of importance scores of the feature.
• Indicating which features are the key for the model
to make its prediction on that instance.
Interpretation
method
Instance
Model
Importance
scores
Making no assumptions on model

Related work : Shapley value
• The Shapley value is an idea in the field of cooperative
game theory.
• This was originally proposed as a characterization of a fair
distribution of a total profit from all the players.
Person:A Person:B Person:C Profit
0 0 1 5
0 1 0 30
0 1 1 50
1 0 0 40
1 0 1 55
1 1 0 75
1 1 1 100
BA C

• The Shapley value of is deﬁned as :
• is the set including all players. (ex. )
• is the function which returns the “proﬁt” by the set .
• considers the contribution of the element .
• means the number of ways for selecting -sized subsets.
i
ϕ(i) =
1
N ∑
S⊆N{i}
1
(
|N| − 1
|S| )
(v(S ∪ {i}) − v(S))
N N = {person A, person B, person C}
v(S) S
v(S ∪ {i}) − v(S) i
(
|N| − 1
|S| )
|S|

Related work : Example
of Shapley value
• The Shapley value of person A is deﬁned as :
ϕ(A) =
1
3 ∑
S⊆{A,B,C}{A}
1
(
|N| − 1
|S| )
(v(S ∪ {A}) − v(S))
=
1
3
(
1
1
(100 − 50) +
1
2
(55 − 5) +
1
2
(75 − 30))
Person:A Person:B Person:C Proﬁt
0 0 1 5
0 1 0 30
0 1 1 50
1 0 0 40
1 0 1 55
1 1 0 75
1 1 1 100

• The Shapley value can be applied to predictive models.
• Each feature is seen as a player in the underlying game.
• Issue : Each evaluation of the Shapley value requires exponential number of
model evaluations. There are two kinds of approaches to deal with the problem.
• Approach1 : Sampling based method
• Randomly sampling feature subsets
• Approach2 : Regression based method
• Sampling feature subsets based on a weighted kernel, and carrying out a
linear regression to estimate Shapley value

Notation
• Feature Vector :
• Note that is the dimension of feature vectors.
• Set of features :
• Sub-vector of features :
• Output variables :
• Output vector of a model given an input vector :
x ∈ 𝒳 ⊂ ℝd
d
S ⊂ {1,2,…, d}
xs = {xj, j ∈ S}
y ∈ 𝒴
x ℙm(Y|x)

Preliminaries : Importance of a feature set
• Here, Importance score of feature set is introduced as:
• Where denotes the expectation over .
• The more similar the prediction produced by to the
prediction produced by , the higher becomes.
S
𝔼m[ ⋅ |x] ℙm( ⋅ |x)
xS
x vx(S)
vx(S) = 𝔼m[
log ℙm(Y|xs) ∣ x
]

Preliminaries : Importance of a feature set
• In many cases, class-specific importance is favored.
• How important is a feature set S to the predicted class?
• Here, following degenerate conditional distribution is
introduced.
• We can then define the importance of a subset S with
respect to using the modified score, which is the expected log
probability of the predicted class.
̂ℙm
̂ℙm(y|x) =
{
1 if y ∈ arg maxy′ℙm(y′|x)
0 otherwise .
vx(S) = ̂𝔼m[
log ℙm(Y|xs) ∣ x
]

Preliminaries : measuring interaction between features
• Consider quantifying the importance of a given -th feature for feature vector .
• A naive way is to compute the importance of set : .
• But it ignores interactions between features.
• For example, when performing sentiment analysis on the following sentence.
This movie is not heartwarming or entertaining.
• Then we wish to quantify the the importance of feature “not”, which plays an
important role in the sentence as being classiﬁed as negative.
• But one would expect that , because “not” itself has neither
negative nor positive sentiment.
i x
{i} Vx({i})
Vx({not}) ≈ 0

Preliminaries : marginal contributions of a feature
• It is essential to consider the interactions of a given feature with other
features.
• A natural way to assess how feature interacts with other features is to
compute difference between the importance of all features in S,
with and without .
• This difference is called marginal contribution of to S, and given by :
• To obtain a simple scaler measure for , we need to aggregate these
marginal contributions over all subsets S including .
• The Shapley value is one way to do so.
i
i
i
i
i
i
mx(S, i) := vx(S) − vx(Si)

Preliminaries : Shapley value
• For k = 1,2,…,d, we let denote the set of k-sized feature subsets
that contain feature .
• The Shapley value is obtained by averaging the marginal
contributions.
• First over the set for a ﬁxed k.
• Then over all possible choices of set size k.
Sk(i)
i
Sk(i)
ϕx(i) =
1
d
d
∑
k=1
1
(
d − 1
k − 1)
∑
S∈Sk(i)
mx(S, i)
Set of k-sized feature sets including feature i

Challenge with computing Shapley value
• The exact computation of the Shapley value leads to
computational difﬁculties.
• We need to calculate marginal contributions for subsets.
• There are some sampling-based approaches to deal with the
problem.
• But such approaches suffer from high variance when the
number of samples to be collected per instance is limited.
2d−1
ϕx(i) =
1
d
d
∑
k=1
1
(
d − 1
k − 1)
∑
S∈Sk(i)
mx(S, i) =
∑
S∋i,S⊆[d]
1
(
d − 1
|S| − 1)
mx(S, i)

Proposed method
Relaxing computational difﬁculties of Shapley-value

Key idea : features can be seen as nodes of a
graph, and they have some relationship.
• In many applications, features can be considered as nodes of a
graph, and we can deﬁne distances between pairs of features
based on the graph structure.
• Features distant in the graph have weak interactions each
other.
• For example, an image is modeled with a grid graph. Pixels
that are far apart may have little effect on each other in the
computation of Shapley value.
• Or, a text is represented as a line graph.
This is A pen

Proposed method : preliminary
• We are given feature vector
• Then we let denote a connected graph
• Each feature is assigned with node .
• Edges represent the interactions between features.
• The graph induces a following distance function on .
• For a given node , its k-neighborhood is the set :
x ∈ ℝd
G = (V, E)
i i
V × V
i ∈ V
𝒩k(i) := {j ∈ V|dG(i, j) ≤ k}
dG(l, m) = the number of edges in shortest path joining l to m .
Gray area is an example of .𝒩2(i)

Proposed method1 : Local-Shapley (L-Shapley)
• Deﬁnition1: Given a model , a sample , and a feature ,
the L-Shapley estimate of order k on a graph is given by :
ℙm x i
G
̂ϕk
x(i) =
1
| 𝒩k(i)| ∑
T∋i,T⊆Nk(i)
1
(
|Nk(i)| − 1
|T| − 1 )
mx(T, i)
k-neighborhoods of i
original Shapley value : ϕx(i) =
∑
S∋i,S⊆[d]
1
(
d − 1
|S| − 1)
mx(S, i)

Proposed method2 : Connected-Shapley (C-Shapley)
• Deﬁnition2: Given a model , a sample , and a feature ,
the C-Shapley estimate of order k on a graph is given by :
where denotes the set of all subsets of that
contain node , and nodes that are connected in .
ℙm x i
G
Ck(i) 𝒩k(i)
i G
̂ϕk
x(i) =
∑
U∈Ck(i)
2
(|U| + 2)(|U| + 1)|U|
mx(U, i)
original Shapley value : ϕx(i) =
∑
S∋i,S⊆[d]
1
(
d − 1
|S| − 1)
mx(S, i)

Examples
• Left subset (blue and red) is summed over in L-Shapely, not
in C-Shapley.
• Right subset (blue and red) is summed over in both L-
Shapley and C-Shapley.

Properties : Error between L-Shapley value
and true Shapley value is upper-bounded.
• S is the subset of k-nearest features of i.
• is the sub-vector having k-nearest features of i.
• is the sub-vector having features not included by S.
XU
XV

Properties : Error between C-Shapley value
and true Shapley value is upper-bounded.
• S is the subset of k-nearest features of i.
• is the connected subset in S.U ∋ i

Experiments : tasks and baselines
• Tasks : image classiﬁcation, and text classiﬁcation
• Baselines : model agnostic methods
• KernelSHAP : regression based approximation of Shapley
• SampleShapley : Random sampling based approximation
of Shapley
• LIME : model agnostic interpretation method linearly
approximating the black-box function around the target
feature.

Experiments : evaluation method
• Evaluation method :
• The change in log-odds scores of the predicted class
before and after masking the top features ranked by
importance scores, where masked words are replaced by
zero paddings.
• Larger decreasing of log-odds means that importance
scores by algorithm could correctly capture importance
of features.

Experiment(1/3) : Text classification
• We study the performances on three neural models and
three datasets:
Dataset Task Method Accuracy
IMDB-Review Sentiment classification Word-based CNN 90.1%
AG news Category classification Character-Based
CNN
90.09%
Yahoo! Answers Category classification LSTM 70.84%

Experiment(1/3) : result
• On IMDB with Word-CNN, the simplest model among the
three, L-Shapley achieves the best performance while LIME,
KernelSHAP and C-Shapley achieve slightly worse
performance.
Better

Experiment(2/3) : Image classiﬁcation
• Datasets :
• A subset of MNIST : Only “3" and “8” are included.
• A subset of CIFAR-10 : Only deers(鹿) and horses(馬) are
included.

Experiment(2/3) : Result
• C-Shapley outperform other methods in both datasets.
Better

Experiment(2/3) : Examples for misclassified
images
• Above image is “3”, and below image is
“8”. They are misclassified into “8”, and
“3”, respectively.
• The masked pixels are colored with red if
activated, (white) and blue otherwise.
• The result seems to show the “reasoning”
of the classifier.

Experiment(3/3) : Evaluation by human
subjects (5 people)
• They use Amazon Mechanical Turk to compare L-Shapley, C-
Shapley and KernelSHAP on IMDB movie reviews (200 reviews).
• Experimental purpose :
• Are humans able to make a decision with top words alone?
• Are humans unable to make a decision with top words masked?
• They ask subjects to classify the sentiment of texts into ﬁve
categories : strongly positive (+2), positive (+1), neutral (0),
negative (-1), strongly negative (-2).

subjects (5 people)
• Texts have three types :
1. raw reviews
2. top 10 words of each review ranked by L-Shapley, C-Shapley and
KernelSHAP
3. reviews with top words being masked
• Masked words are produced by the L-Shapley, C-Shapley, and
KernelSHAP until the probability score of the correct class produced by
the model is lower than 10%.
• Around 14.6% of words in each review are masked for L-Shapley and C-
Shapley, and 31.6% for KernelSHAP.

subjects (5 people)
• Evaluation metrics :
• Consistency (0 or 1) between true labels and labels from
human subjects.
• Standard deviation of scores on each reviews
• This is as a measure of disagreement between humans.
• The absolute value of the averaged scores.
• This is as a measure of conﬁdence of decision.

Experiment(3/3) : result
• Humans become more consistent and conﬁdent when they are presented
with top words. On the other hand, when top words are masked, humans are
easier to make mistakes and are less certain.
• C-Shapley yields the highest performance in terms of consistency, agreement,
and conﬁdence.
• L-Shapley harms the human judgement the most among the three algorithms.

Conclusion
• They have proposed two algorithms; L-Shapley and C-
Shapley for instance-wise and model-agnostic
interpretation.
• They demonstrated the superior performance of these
algorithms compared to other methods on black-box
models in both text and image classiﬁcation with both
quantitative metrics and human evaluation.

[Paper reading] L-SHAPLEY AND C-SHAPLEY: EFFICIENT MODEL INTERPRETATION FOR STRUCTURED DATA

More Related Content

What's hot (20)

Similar to [Paper reading] L-SHAPLEY AND C-SHAPLEY: EFFICIENT MODEL INTERPRETATION FOR STRUCTURED DATA (20)

More from Daiki Tanaka (12)

Recently uploaded (20)

[Paper reading] L-SHAPLEY AND C-SHAPLEY: EFFICIENT MODEL INTERPRETATION FOR STRUCTURED DATA