Incorporating Priors with Feature Attribution on Text Classification

Abstract

Feature attribution methods, proposed recently, help users interpret the predictions of complex models. Our approach integrates feature attributions into the objective function to allow machine learning practitioners to incorporate priors in model building. To demonstrate the effectiveness our technique, we apply it to two tasks: (1) mitigating unintended bias in text classifiers by neutralizing identity terms; (2) improving classifier performance in a scarce data setting by forcing the model to focus on toxic terms. Our approach adds an L2 distance loss between feature attributions and task-specific prior values to the objective. Our experiments show that i) a classifier trained with our technique reduces undesired model biases without a trade off on the original task; ii) incorporating priors helps model performance in scarce data settings.

Accepted to ACL 2019

References in corpus (11)

• Axiomatic Attribution for Deep Networks
• Learning Important Features Through Propagating Activation Differences
• European Union regulations on algorithmic decision-making and a "right to explanation"
• Equality of Opportunity in Supervised Learning
• Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
• Understanding Black-box Predictions via Influence Functions
• SmoothGrad: removing noise by adding noise
• Sanity Checks for Saliency Maps
• Consistent Individualized Feature Attribution for Tree Ensembles
• Distilling a Neural Network Into a Soft Decision Tree
• Pathologies of Neural Models Make Interpretations Difficult

Cited by in corpus (6)

• EDITS: Modeling and Mitigating Data Bias for Graph Neural Networks
• Explaining Explanations: Axiomatic Feature Interactions for Deep Networks
• Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
• Estimating and Improving Fairness with Adversarial Learning
• Fair Hate Speech Detection through Evaluation of Social Group Counterfactuals
• Corpus-level and Concept-based Explanations for Interpretable Document Classification