Corpus ID: 13157581

Parallel and Distributed Block-Coordinate Frank-Wolfe Algorithms

@inproceedings{Wang2016ParallelAD,
  title={Parallel and Distributed Block-Coordinate Frank-Wolfe Algorithms},
  author={Yu-Xiang Wang and Veeranjaneyulu Sadhanala and Wei Dai and W. Neiswanger and S. Sra and E. Xing},
  booktitle={ICML},
  year={2016}
}
  • Yu-Xiang Wang, Veeranjaneyulu Sadhanala, +3 authors E. Xing
  • Published in ICML 2016
  • Computer Science, Mathematics
  • We study parallel and distributed Frank-Wolfe algorithms; the former on shared memory machines with mini-batching, and the latter in a delayed update framework. In both cases, we perform computations asynchronously whenever possible. We assume block-separable constraints as in Block-Coordinate Frank-Wolfe (BCFW) method (Lacoste-Julien et al., 2013), but our analysis subsumes BCFW and reveals problemdependent quantities that govern the speedups of our methods over BCFW. A notable feature of our… CONTINUE READING
    29 Citations
    Distributing Frank-Wolfe via Map-Reduce
    Distributing Frank–Wolfe via map-reduce
    Cyclades: Conflict-free Asynchronous Machine Learning
    • 50
    • PDF
    Communication-Efficient Projection-Free Algorithm for Distributed Optimization
    • 2
    • PDF
    Asynchronous Parallel Coordinate Minimization for MAP Inference
    • 5
    • PDF
    Distributed Asynchronous Optimization with Unbounded Delays: How Slow Can You Go?
    • 18
    • PDF
    Randomized Block Frank–Wolfe for Convergent Large-Scale Learning
    • 11
    • PDF
    One Sample Stochastic Frank-Wolfe
    • 12
    • PDF
    Quantized Frank-Wolfe: Communication-Efficient Distributed Optimization
    • 5
    • PDF

    References

    SHOWING 1-10 OF 56 REFERENCES
    Hogwild: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent
    • 1,722
    • Highly Influential
    • PDF
    Parallel coordinate descent methods for big data optimization
    • 389
    • PDF
    A Distributed Frank-Wolfe Algorithm for Communication-Efficient Sparse Learning
    • 31
    • PDF
    Petuum: A Framework for Iterative-Convergent Distributed ML
    • 31
    • PDF
    More Effective Distributed ML via a Stale Synchronous Parallel Parameter Server
    • 442
    • PDF
    New analysis and results for the Frank–Wolfe method
    • 91
    • PDF
    On the Global Linear Convergence of Frank-Wolfe Optimization Variants
    • 233
    • PDF
    Petuum: A New Platform for Distributed Machine Learning on Big Data
    • 178
    • PDF