Simultaneous Unlearning of Multiple Protected User Attributes From Variational Autoencoder Recommenders Using Adversarial Training
This work addresses fairness and privacy concerns in recommender systems for users by enabling simultaneous removal of multiple protected attributes, though it is incremental as it builds on prior single-attribute unlearning methods.
The paper tackles the problem of neural collaborative filtering models inadvertently encoding users' protected attributes like gender and age into latent embeddings, which can lead to unfair treatment and privacy issues. The proposed AdvXMultVAE method simultaneously removes multiple protected attributes using adversarial training, achieving better bias mitigation and anonymity than single-attribute removal approaches on music and movie datasets.
In widely used neural network-based collaborative filtering models, users' history logs are encoded into latent embeddings that represent the users' preferences. In this setting, the models are capable of mapping users' protected attributes (e.g., gender or ethnicity) from these user embeddings even without explicit access to them, resulting in models that may treat specific demographic user groups unfairly and raise privacy issues. While prior work has approached the removal of a single protected attribute of a user at a time, multiple attributes might come into play in real-world scenarios. In the work at hand, we present AdvXMultVAE which aims to unlearn multiple protected attributes (exemplified by gender and age) simultaneously to improve fairness across demographic user groups. For this purpose, we couple a variational autoencoder (VAE) architecture with adversarial training (AdvMultVAE) to support simultaneous removal of the users' protected attributes with continuous and/or categorical values. Our experiments on two datasets, LFM-2b-100k and Ml-1m, from the music and movie domains, respectively, show that our approach can yield better results than its singular removal counterparts (based on AdvMultVAE) in effectively mitigating demographic biases whilst improving the anonymity of latent embeddings.