Making Neural Networks Interpretable with Attribution: Application to Implicit Signals Prediction

Abstract

Explaining recommendations enables users to understand whether recommended items are relevant to their needs and has been shown to increase their trust in the system. More generally, if designing explainable machine learning models is key to check the sanity and robustness of a decision process and improve their efficiency, it however remains a challenge for complex architectures, especially deep neural networks that are often deemed "black-box". In this paper, we propose a novel formulation of interpretable deep neural networks for the attribution task. Differently to popular post-hoc methods, our approach is interpretable by design. Using masked weights, hidden features can be deeply attributed, split into several input-restricted sub-networks and trained as a boosted mixture of experts. Experimental results on synthetic data and real-world recommendation tasks demonstrate that our method enables to build models achieving close predictive performances to their non-interpretable counterparts, while providing informative attribution interpretations.

14th ACM Conference on Recommender Systems (RecSys '20)

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• A Unified Approach to Interpreting Model Predictions
• The information bottleneck method
• Learning Important Features Through Propagating Activation Differences
• SmoothGrad: removing noise by adding noise
• The composite absolute penalties family for grouped and hierarchical variable selection
• NAIS: Neural Attentive Item Similarity Model for Recommendation
• Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches
• Learning with Structured Sparsity
• RecVAE: a New Variational Autoencoder for Top-N Recommendations with Implicit Feedback
• Molecule Attention Transformer
• High-Dimensional Non-Linear Variable Selection through Hierarchical Kernel Learning
• Restricting the Flow: Information Bottlenecks for Attribution
• RPGAN: GANs Interpretability via Random Routing
• Modelling Sequential Music Track Skips using a Multi-RNN Approach
• Session-based Sequential Skip Prediction via Recurrent Neural Networks
• Sequential Skip Prediction with Few-shot in Streamed Music Contents

Cited by in corpus (5)

• Explainability in Music Recommender Systems
• Why People Skip Music? On Predicting Music Skips using Deep Reinforcement Learning
• Towards Rigorous Interpretations: a Formalisation of Feature Attribution
• Searching Personal Collections
• Contributions to Representation Learning with Graph Autoencoders and Applications to Music Recommendation