Individual Conditional Expectation Plots

Description

ICE plots display the predicted output for individual instances as a function of a feature, with all other features held fixed for each instance. Each line on an ICE plot represents one instance's prediction trajectory as the feature of interest changes, revealing whether different instances are affected differently by that feature.

Example Use Cases

Explainability

Examining how house price predictions vary with property age for individual properties, revealing that whilst most houses follow a declining price trend with age, historic properties (built before 1900) show different patterns due to heritage value.

Analysing how individual patients' diabetes risk predictions change with BMI, showing that whilst most patients follow the expected increasing risk pattern, some patients with specific genetic markers show different response curves.

Limitations

Plots can become cluttered and difficult to interpret when displaying many instances simultaneously.
Does not provide automatic summarisation of overall effects, requiring manual visual inspection to identify patterns.
Still assumes all other features remain fixed at their observed values, which may not reflect realistic scenarios.
Cannot reveal interactions between the plotted feature and other features for individual instances.

Resources

Peeking Inside the Black Box: Visualizing Statistical Learning with Plots of Individual Conditional Expectation

Research Paper•Alex Goldstein et al.•Sep 25, 2013

Bringing a Ruler Into the Black Box: Uncovering Feature Impact from Individual Conditional Expectation Plots

Research Paper•Andrew Yeh and Anhthy Ngo•Sep 6, 2021

Explainable AI(XAI) - A guide to 7 packages in Python to explain ...

Tutorial

Communicating Uncertainty in Machine Learning Explanations: A Visualization Analytics Approach for Predictive Process Monitoring

Research Paper•Nijat Mehdiyev, Maxim Majlatow, and Peter Fettke

Related Techniques

Name	Description	Assurance Goals
Model Pruning	Model pruning systematically removes less important weights, neurons, or entire layers from neural networks to create smaller, more efficient models whilst maintaining performance. This process involves iterative removal based on importance criteria (weight magnitudes, gradient information, activation patterns) followed by fine-tuning. Pruning can be structured (removing entire neurons/channels) or unstructured (removing individual weights), with structured pruning providing greater computational benefits and interpretability through simplified architectures.	Explainability Reliability Safety
Saliency Maps	Saliency maps are visual explanations for image classification models that highlight which pixels in an image most strongly influence the model's prediction. Computed by calculating gradients of the model's output with respect to input pixels, saliency maps produce heatmaps where brighter regions indicate pixels that, when changed, would most significantly affect the prediction. This technique helps users understand which parts of an image the model is 'looking at' when making decisions.	Explainability Fairness
Taylor Decomposition	Taylor Decomposition is a mathematical technique that explains neural network predictions by computing first-order and higher-order derivatives of the network's output with respect to input features. It decomposes the prediction into relevance scores that indicate how much each input feature and feature interaction contributes to the final decision. The method uses Layer-wise Relevance Propagation (LRP) principles to trace prediction contributions backwards through the network layers, providing precise mathematical attributions for each input element.	Explainability
Principal Component Analysis	Principal Component Analysis transforms high-dimensional data into a lower-dimensional representation by finding the directions (principal components) that capture the maximum variance in the data. Each component is a linear combination of original features, with the first component explaining the most variance, the second component the most remaining variance orthogonal to the first, and so on. This technique reveals underlying patterns in data structure, enables visualization of complex datasets, and helps identify which combinations of features drive the most variation in the data.	Explainability
Permutation Importance	Permutation Importance quantifies a feature's contribution to a model's performance by randomly shuffling its values and measuring the resulting drop in predictive accuracy. If shuffling a feature significantly degrades the model's performance, that feature is considered important. This model-agnostic technique helps identify which inputs are genuinely driving predictions, rather than just being correlated with the outcome.	Explainability Reliability
Monte Carlo Dropout	Monte Carlo Dropout estimates prediction uncertainty by applying dropout (randomly setting neural network weights to zero) during inference rather than just training. It performs multiple forward passes through the network with different random dropout patterns and collects the resulting predictions to form a distribution. Low variance across predictions indicates epistemic certainty (the model is confident), while high variance suggests epistemic uncertainty (the model is unsure). This technique transforms any dropout-trained neural network into a Bayesian approximation for uncertainty quantification.	Explainability Reliability