Geometry Aware Mappings for High Dimensional Sparse Factors
This work addresses efficiency issues in large-scale recommendation systems, though it appears incremental as it builds on existing factorization models with optimization techniques.
The paper tackles the problem of high computational cost in real-time matrix factorization for recommendation and search by introducing a framework that uses geometry-aware mappings to create sparse embeddings, achieving faster runtime with minimal accuracy loss.
While matrix factorisation models are ubiquitous in large scale recommendation and search, real time application of such models requires inner product computations over an intractably large set of item factors. In this manuscript we present a novel framework that uses the inverted index representation to exploit structural properties of sparse vectors to significantly reduce the run time computational cost of factorisation models. We develop techniques that use geometry aware permutation maps on a tessellated unit sphere to obtain high dimensional sparse embeddings for latent factors with sparsity patterns related to angular closeness of the original latent factors. We also design several efficient and deterministic realisations within this framework and demonstrate with experiments that our techniques lead to faster run time operation with minimal loss of accuracy.