Robust Aggregation for Federated Sequential Recommendation with Sparse and Poisoned Data

TL;DR

This paper introduces a robust aggregation framework for federated sequential recommendation systems that effectively handles sparse user data and malicious client updates, enhancing privacy and model reliability.

Contribution

It proposes a novel defence-aware aggregation method with representation constraints and sequence regularisation tailored for federated sequential recommendation under adversarial and sparse data conditions.

Findings

Improved robustness against poisoned client updates

Enhanced stability of user and item embeddings

Maintained temporal coherence in recommendations

Abstract

Federated sequential recommendation distributes model training across user devices so that behavioural data remains local, reducing privacy risks. Yet, this setting introduces two intertwined difficulties. On the one hand, individual clients typically contribute only short and highly sparse interaction sequences, limiting the reliability of learned user representations. On the other hand, the federated optimisation process is vulnerable to malicious or corrupted client updates, where poisoned gradients can significantly distort the global model. These challenges are particularly severe in sequential recommendation, where temporal dynamics further complicate signal aggregation. To address this problem, we propose a robust aggregation framework tailored for federated sequential recommendation under sparse and adversarial conditions. Instead of relying on standard averaging, our method introduces a defence-aware aggregation mechanism that identifies and down-weights unreliable client updates while preserving informative signals from sparse but benign participants. The framework incorporates representation-level constraints to stabilise user and item embeddings, preventing poisoned or anomalous contributions from dominating the global parameter space. In addition, we integrate sequence-aware regularisation to maintain temporal coherence in user modelling despite limited local observations.