Turbo-CF: Matrix Decomposition-Free Graph Filtering for Fast Recommendation
This addresses the need for fast recommendations in practical scenarios by reducing computational costs, though it is incremental as it builds on existing graph filtering methods.
The paper tackles the computational inefficiency of graph filtering-based collaborative filtering methods by proposing Turbo-CF, a training-free and matrix decomposition-free approach that uses polynomial graph filters, achieving runtime under 1 second on real-world datasets with accuracy comparable to state-of-the-art competitors.
A series of graph filtering (GF)-based collaborative filtering (CF) showcases state-of-the-art performance on the recommendation accuracy by using a low-pass filter (LPF) without a training process. However, conventional GF-based CF approaches mostly perform matrix decomposition on the item-item similarity graph to realize the ideal LPF, which results in a non-trivial computational cost and thus makes them less practical in scenarios where rapid recommendations are essential. In this paper, we propose Turbo-CF, a GF-based CF method that is both training-free and matrix decomposition-free. Turbo-CF employs a polynomial graph filter to circumvent the issue of expensive matrix decompositions, enabling us to make full use of modern computer hardware components (i.e., GPU). Specifically, Turbo-CF first constructs an item-item similarity graph whose edge weights are effectively regulated. Then, our own polynomial LPFs are designed to retain only low-frequency signals without explicit matrix decompositions. We demonstrate that Turbo-CF is extremely fast yet accurate, achieving a runtime of less than 1 second on real-world benchmark datasets while achieving recommendation accuracies comparable to best competitors.