🤖 AI Summary
This work addresses the challenge of efficiently modeling long user behavioral sequences under stringent memory and latency constraints. The authors propose HoloMambaRec, a novel architecture that integrates holographic reduced representations with selective state space models (Mamba). By employing holographic embeddings to unify item and attribute information and leveraging a shallow Mamba encoder, the model achieves linear-time sequence modeling and constant-time inference, enabling temporal bundling and inference compression. Experimental results demonstrate that HoloMambaRec matches the performance of GRU4Rec and outperforms SASRec on Amazon Beauty and MovieLens-1M datasets, while substantially reducing memory consumption—making it particularly suitable for resource-constrained scenarios with as few as 10 training epochs.
📝 Abstract
Sequential recommender systems must model long-range user behavior while operating under strict memory and latency constraints. Transformer-based approaches achieve strong accuracy but suffer from quadratic attention complexity, forcing aggressive truncation of user histories and limiting their practicality for long-horizon modeling. This paper presents HoloMambaRec, a lightweight sequential recommendation architecture that combines holographic reduced representations for attribute-aware embedding with a selective state space encoder for linear-time sequence processing. Item and attribute information are bound using circular convolution, preserving embedding dimensionality while encoding structured metadata. A shallow selective state space backbone, inspired by recent Mamba-style models, enables efficient training and constant-time recurrent inference. Experiments on Amazon Beauty and MovieLens-1M under a 10-epoch budget show that HoloMambaRec surpasses SASRec on both datasets, attains state-of-the-art ranking on MovieLens-1M, and trails only GRU4Rec on Amazon Beauty, all while maintaining substantially lower memory complexity. The design further incorporates forward-compatible mechanisms for temporal bundling and inference-time compression, positioning HoloMambaRec as a practical and extensible alternative for scalable, metadata-aware sequential recommendation.