A Logical-Rule Autoencoder for Interpretable Recommendations
This addresses the need for transparency and accountability in recommendation systems, particularly in applications requiring interpretability, though it is an incremental improvement focused on a specific domain.
The paper tackles the problem of black-box deep learning recommendation models by proposing a Logical-rule Interpretable Autoencoder (LIA) that learns explicit, human-readable rules for collaborative filtering, achieving improved recommendation performance over traditional baselines while maintaining full interpretability.
Most deep learning recommendation models operate as black boxes, relying on latent representations that obscure their decision process. This lack of intrinsic interpretability raises concerns in applications that require transparency and accountability. In this work, we propose a Logical-rule Interpretable Autoencoder (LIA) for collaborative filtering that is interpretable by design. LIA introduces a learnable logical rule layer in which each rule neuron is equipped with a gate parameter that automatically selects between AND and OR operators during training, enabling the model to discover diverse logical patterns directly from data. To support functional completeness without doubling the input dimensionality, LIA encodes negation through the sign of connection weights, providing a parameter-efficient mechanism for expressing both positive and negated item conditions within each rule. By learning explicit, human-readable reconstruction rules, LIA allows users to directly trace the decision process behind each recommendation. Extensive experiments show that our method achieves improved recommendation performance over traditional baselines while remaining fully interpretable. Code and data are available at https://github.com/weibowen555/LIA.