Matrix Factorization on GPUs with Memory Optimization and Approximate Computing

@article{Tan2018MatrixFO,
  title={Matrix Factorization on GPUs with Memory Optimization and Approximate Computing},
  author={W. Tan and S. Chang and L. Fong and Cheng Li and Zijun Wang and L. Cao},
  journal={Proceedings of the 47th International Conference on Parallel Processing},
  year={2018}
}
  • W. Tan, S. Chang, +3 authors L. Cao
  • Published 2018
  • Computer Science
  • Proceedings of the 47th International Conference on Parallel Processing
  • Matrix factorization (MF) discovers latent features from observations, which has shown great promises in the fields of collaborative filtering, data compression, feature extraction, word embedding, etc. [...] Key Method The former exploits GPU memory hierarchy to increase data reuse, while the later reduces unneccessary computing without hurting the convergence of learning algorithms. Extensive experiments on large-scale datasets show that our solution not only outperforms the competing CPU solutions by a large…Expand Abstract
    Block based Singular Value Decomposition approach to matrix factorization for recommender systems
    • 4
    • PDF
    BMF: Block matrix approach to factorization of large scale data
    • 1
    • PDF
    Identifying and (automatically) remedying performance problems in CPU/GPU applications
    Diogenes: looking for an honest CPU/GPU performance measurement tool
    • 2
    • Highly Influenced

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 20 REFERENCES
    Fast and Robust Parallel SGD Matrix Factorization
    • 36
    • Highly Influential
    • PDF
    Parallel CCD++ on GPU for Matrix Factorization
    • 14
    • Highly Influential
    Distributed Matrix Completion
    • 126
    • Highly Influential
    • PDF
    A fast parallel SGD for matrix factorization in shared memory systems
    • 188
    • Highly Influential
    • PDF
    Large-scale matrix factorization with distributed stochastic gradient descent
    • 597
    • Highly Influential
    • PDF
    Scalable Coordinate Descent Approaches to Parallel Matrix Factorization for Recommender Systems
    • 248
    • Highly Influential
    • PDF
    Accelerating collaborative filtering using concepts from high performance computing
    • 31
    • Highly Influential
    • PDF
    Hogwild: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent
    • 1,686
    • Highly Influential
    • PDF
    Factorbird - a Parameter Server Approach to Distributed Matrix Factorization
    • 27
    • Highly Influential
    • PDF
    Exploiting Bounded Staleness to Speed Up Big Data Analytics
    • 114
    • Highly Influential
    • PDF