Questions tagged [shapley-value]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
0 votes
0 answers
6 views

Interpreting Shapley Values on Breast Cancer

I was analyzing Shapley Values on the Wisconsin breast cancer data set (binary classification). I applied it on Random Forest and on Ridge and Lasso Regression. However the summary plot seems to be ...
user avatar
  • 1
0 votes
0 answers
11 views

Assessing importance of interactions between categorical features

The issues with using feature_importance of models such as XGBoost, or even using packages like SHAP or ELI5, is that the results are displayed in a way that doesn'...
user avatar
  • 519
0 votes
0 answers
12 views

Post analysis using raw data or SHAP values in Machine learning

Let's say I have SHAP value returned in dataframe for input variables like below ...
user avatar
  • 1,780
1 vote
0 answers
27 views

Should we prefer mean of absolute or square Shapley values?

Usually (for linear models at least) Shapley values are defined as deviations to some mean. By default, it seems the bar plot of the popular python package shap diplays the mean of absolute Shapley ...
user avatar
  • 111
0 votes
1 answer
34 views

Can SHAP be used for linear mixed models?

Can SHAP importance be used for linear mixed models? I've seen it used for a variety of different modeling methods and was curious if it was possible to use it for linear mixed models? I am using the ...
user avatar
  • 39
1 vote
0 answers
25 views

Appropriateness of using SHAP values to evaluate a model

I am a deep learning researcher that would like to use SHAP values to assess the relative importance of input features on the model's final score. Colleagues of mine have taken issue with the method ...
user avatar
  • 21
0 votes
0 answers
16 views

Why does the slow version of TreeExplainer algorithm have a complexity of $TLM2^M$

I'm trying to understand the TreeExplainer algorithms better from From Local Explanations to Global Understanding with Explainable AI for Trees, but confused by how the time complexity is derived even ...
user avatar
  • 887
0 votes
0 answers
30 views

Model average prediction - Usefulness and interpretation

I am working on binary classification problem using random forest with a dataset shape of 977,6. class proportion is 77:23. This post is born out these two posts here and here I recently ran the SHAP ...
user avatar
  • 1,780
1 vote
1 answer
117 views

Why SHAP base/expected value is 0.5 for all my instances?

I am working on a binary classification using random forest model, neural networks in which am using SHAP to explain the model predictions. I followed the tutorial and wrote the below code to get the ...
user avatar
  • 1,780
3 votes
1 answer
170 views

What is the use of expected value in machine learning models?

I see that we have a concept called expected value being used in machine learning (ML) models. For example, SHAP has a concept called Expected value. It means when ...
user avatar
  • 1,780
1 vote
1 answer
48 views

SHAP values and feature-target correlations contradict each other - why?

I am using SHAP values in a model explainability analysis and I am seeing a pattern of results where the SHAP values are clearly pointing in a positive or negative direction for most features, i.e. ...
user avatar
  • 41
0 votes
0 answers
28 views

Shapley values for the three clusters by cluster number KMeans algorithm

I am trying to replicate this https://cast42.github.io/blog/datascience/python/clustering/altair/shap/2020/04/23/explain-clusters-to-business.html#Kmeans-clustering But using R and not Python as in ...
user avatar
  • 1
0 votes
0 answers
15 views

How to calculate Shapley Value when the order doesn't matter?

I'm trying to use Shapley Value to evaluate the importance of models in federated learning. Let's take two clients A, B as example. When no one joins, the initial value is ...
user avatar
0 votes
2 answers
31 views

How to calculate SHAP for a factor in a linear model?

Shapley additive explanation (SHAP) are used to explain the prediction of a model $Y = f(X_1,...,X_p) + \varepsilon$. If we observe $x_1,...,x_p$ and predict $y$, then for each $i$ the contribution of ...
user avatar
0 votes
0 answers
73 views

Interpreting SHAP interaction values

I have a trained model where gender, var_a, var_b, .. , var_g are binary features and an age variable too. Gender = 0 : male and Gender = 1 female. Similarly var_a and other features are binary. While ...
user avatar
0 votes
0 answers
27 views

beeswarm plot in SHAP: why does sample size vary across features?

Take this plot, for instance: Why so many dots for daily_time_spent_onsite but only a few dots for male? If every dot is the ...
user avatar
  • 135
0 votes
0 answers
36 views

Interactions in SHAP Values

I am interested in interactions in SHAP values. For example, how the SHAP value of one feature can be high or low depending on the value of another feature. I was initially interested in this question ...
user avatar
  • 173
0 votes
0 answers
35 views

Majority of SHAP values generated via SuperPixels are negative

I built a neural network model for predicting the demand for taxi trips in NYC, which receives a normalized input - previous demand for the last two hours of shape (4,20,20) - like and produces an ...
user avatar
  • 533
1 vote
0 answers
155 views

SHAP for stacking classifier

We are using a stacking classifier to solve a classification problem. The data feed 5 base models, the predicted probabilities of the base models feed the supervisory classifier. We would like to use ...
user avatar
4 votes
2 answers
77 views

When interpreting machine learning models, should preprocessing steps be considered as part of "model"?

Suppose I have some inputs on which I first apply some feature engineering and then apply a machine learning algorithm such as random forest to make predictions. Now, if I want to interpret/explain ...
user avatar
6 votes
0 answers
262 views

How shap values behave in terms of multicollinearity in Trees, Ensemble, GradientBoosting and GAM/Boosting

I set up an experiment with these 8 Regressor Methods: sklearn package DecisionTreeRegressor, RandomForestRegressor, ExtraTreesRegressor, GradientBoostingRegressor other packages CatBoostRegressor ,...
user avatar
0 votes
0 answers
126 views

Does Kernel SHAP assume feature independence?

I'm studying SHAP values for the evaluation of feature importance of deep learning models. However, different sources have different description of the method. In Telesens, the author writes "...
user avatar
  • 91
1 vote
0 answers
28 views

Why do Shapley values increase over time?

I calculated the Shapley values (using xgboot package, gbm regression model) of several big actors in the cocoa market and received results which I cannot explain: it seems that Shapley increases (the ...
user avatar
6 votes
0 answers
823 views

Are SHAP values potentially misleading when predictors are highly correlated?

Are SHAP (SHapley Additive exPlanations) values potentially misleading when predictors are highly correlated? How and why? If so, is there any guidance on when not to use SHAP? Are there any rules ...
user avatar
  • 3,768
0 votes
0 answers
33 views

Calculate Shap Values for incremental/online model

My task is to calculate SHAP values for the samples. ...
user avatar
1 vote
2 answers
65 views

feature importance aggregation

I have more of a conceptual question I was hoping to get some feedback on. I am trying to run a boosted regression ML model to identify a subset of important predictors for some clinical condition. ...
user avatar
  • 107
5 votes
1 answer
446 views

Is there any reason to use LIME now that shap is available?

The context: explaining a binary classifier XGBoost model. If we say that we are limited to the LIME and Shapley Additive Explanation aka "shap" package, is there any reason to use LIME? My ...
user avatar
  • 453
0 votes
0 answers
26 views

Aggregating SHAP values obtained from different models?

I have SHAP values for two models (sklearns's GradientBoostingClassifier and RandomForestClassifier). They are positively correlated. Rank correlation (r_s) between the SHAP values of the two models ...
user avatar
2 votes
0 answers
32 views

shap value of a shuffled feature [closed]

Imagine I have a model to predict a 100 target values y based on 10 features stored in a X table (100 by 10). Similarly to what ...
user avatar
  • 21
1 vote
0 answers
50 views

How does Shapley obtain values in the probability space in tree classifiers?

By default, the Shapley values for a tree explainer (e.g. based on xgboost) are in the log odds space (where they are additive). However, there's a functionality in the package to obtain values in the ...
user avatar
  • 121
1 vote
1 answer
48 views

Can adding more data, influence the SHAPley values?

I am having a classification dataset. Everytime I add more data to the dataset the SHAP values are getting changed and the top list of features are changed. Can anyone explain what is this behavior? ...
user avatar
  • 115
1 vote
0 answers
50 views

Calculate overall attribution with Shapley values

I'm using a shap KernelExplainer to interprete my model and I'd like to display an indication of why the explainer sorted the features in that order. For example, some kind of overall score that ...
user avatar
1 vote
1 answer
376 views

Shap values on scaled dataset

I am working on a binary classification problem for heart disease prediction. I have scaled the dataset using Standard Scaler and I am trying to understand the model generated using SHAP values ...
user avatar
5 votes
1 answer
3k views

Can I scale and then interpret shap values as percent contribution to the prediction?

Let's say I have prediction for an observation with 3 shap values: -2, 3 and 5 for feature A, B and C respectively. Then I scale the absolute value of the shap values so they sum to 1 (i.e A=0.2, B=0....
user avatar
  • 53
3 votes
1 answer
58 views

Carrying Out Interventions Based on ML "Feature Importances"

Recently, I have been studying causal inference and have come to a bit of a crossroads with respect to making decisions based on the analysis of data (especially in a business/industry setting). ...
user avatar
  • 667
2 votes
0 answers
31 views

Are SHAP feature importance measures biased towards variables with many potential cut points?

Variable importance measures in Random Forests can be biased towards variables with more categories, even using permutation-based methods: https://bmcbioinformatics.biomedcentral.com/articles/10.1186/...
user avatar
  • 21
2 votes
0 answers
173 views

What are SHAP loss values?

Could anybody give a simple explanation of SHAP loss values? Perhaps with an example? (I googled it but didn't find a lot and the things I found were not very clear... at least not to me)
user avatar
  • 5,916
6 votes
0 answers
1k views

Interpreting SHAP Dependence Plot for Categorical Variables

I'm reading about the use of Shapley values for explaining complex machine learning models and I'm confused about how I should interpret the SHAP independence plot in the case of a categorical ...
user avatar
1 vote
0 answers
170 views

Using SHAP or LIME to explain new predictions?

I've previously used SHAP and LIME to explain predictions from a training set, i.e. I have the actual target value. Is it possible to do the same to explain new predictions, i.e. I don't have the ...
user avatar
1 vote
1 answer
112 views

Time varying Shapley Decomposition

I have read a lot on Shapley decompositions for relative contributions of regressors in linear regression. I was wondering if there is a way to do time-varying relative contributions. For example: If ...
user avatar
1 vote
0 answers
277 views

Shapley values do not add up to prediction

I am building a RF model to predict cervical cancer from biopsies (data). I am then trying to explain the predictions using Shapley values. To do this I use the mli ...
user avatar
  • 488
0 votes
2 answers
397 views

Feature explanation with SHAP

I need to have a feature explanation of a predictor. I've read about SHAP values, but it seems I have to make coalitions with the features and make a prediction with every one of them. I have $100$ ...
user avatar
  • 1
1 vote
0 answers
25 views

Component contributions in Additive Model Time Series

I have trained a model for forecasting time series in a greedy procedure: Fit the Trend component T(t) of the series on the original signal y(t) Fit a Cyclical/Seasonal S(t) component of the series ...
user avatar
1 vote
1 answer
255 views

Can we use Shap to interpret output changes?

Can we calculate the difference between Shapley values to interpret changes in the output? More precisely, if we get Shapley values for two different inputs, can we compare them to understand how much ...
user avatar
  • 147
1 vote
0 answers
91 views

Scaling baseline and SHAPs back to original class rate

I have an imbalanced dataset (positive class rate = 1%) and have downsampled the negative class to give me a 50/50 balance in the two classes. The outputs from this model look adequate. Ignoring the ...
user avatar
  • 131
5 votes
2 answers
712 views

Reasons that LIME and SHAP might not agree with intuition

I'm leveraging the Python packages lime and shap to explain single (test-set) predictions that a basic, trained model is making on new, tabular data. WLOG, the explanations generated by both methods ...
user avatar
3 votes
1 answer
969 views

Best way to assess SHAP values variability

Let's say you don't have a lot of data to fit a model but you still want to have a sense of feature importance for your model. SHAP values are a very interesting tool to do that. Not having a lot of ...
user avatar
  • 263
8 votes
0 answers
840 views

Difference between Shapley values and SHAP

The Paper regarding die SHAP value gives a formula for the Shapley Values in (4) and for SHAP values apparently (?) in (8) Still I dont really understand the difference between Shapley and SHAP ...
user avatar
  • 203
2 votes
0 answers
706 views

Shapley value vs ridge regression

My goal is to get the feature importance for multiple regression. I have a data set with some multicollinearity. I found two methods to solve this problem. The first one is the Shapley value. ...
user avatar
  • 245
0 votes
2 answers
988 views

How to perform SHAP explainer on a system of models

I have developed a model with Autoencoder + XGBoost. Autoencoder is used to reduce dimensionality and then passed on to XGBoost for prediction. I would like to understand the feature importance of the ...
user avatar