Interpretability

Edit

General • 16 methods

Interpretability Methods seek to explain the predictions made by neural networks by introducing mechanisms to enduce or enforce interpretability. For example, LIME approximates the neural network with a locally interpretable model. Below you can find a continuously updating list of interpretability methods.

Methods

Add a Method

Method	Year	Papers
SHAP A Unified Approach to Interpreting Model Predictions	2017	326
LIME "Why Should I Trust You?": Explaining the Predictions of Any Classifier	2016	276
CAM Is Object Localization for Free? - Weakly-Supervised Learning With Convolutional Neural Networks	2015	206
Monte Carlo Dropout Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning	2015	147
Visual Analytics Visual Analytics of Multivariate Networks with Representation Learning and Composite Variable Construction	2023	39
NAM Neural Additive Models: Interpretable Machine Learning with Neural Nets	2020	12
Network Dissection Interpreting Deep Visual Representations via Network Dissection	2017	11
CDEP Interpretations are useful: penalizing explanations to align neural networks with prior knowledge	2019	2
Universal Probing Universal and Independent: Multilingual Probing Framework for Exhaustive Model Interpretation and Evaluation	2022	2
Agglomerative Contextual Decomposition Hierarchical interpretations for neural network predictions	2018	1
Disentangled Attribution Curves Disentangled Attribution Curves for Interpreting Random Forests and Boosted Trees	2019	1
Symbolic Deep Learning Discovering Symbolic Models from Deep Learning with Inductive Biases	2020	1
Syntax Heat Parse Tree KERMIT: Complementing Transformer Architectures with Encoders of Explicit Syntactic Interpretations		1
Hierarchical Network Dissection Interpreting Face Inference Models using Hierarchical Network Dissection	2021	1
TE2Rules TE2Rules: Explaining Tree Ensembles using Rules	2022	1
DecomCAM Decom--CAM: Tell Me What You See, In Details! Feature-Level Interpretation via Decomposition Class Activation Map	2023	1

Interpretability Edit

Methods Add a Method

Interpretability

Edit

Methods

Add a Method