18265

Fast Locality Sensitive Hashing for Beam Search on GPU

Xing Shi, Shizhen Xu, Kevin Knight
Department of Computer Science, University of Southern California
arXiv:1806.00588 [cs.CL], (2 Jun 2018)

@article{shi2018fast,

   title={Fast Locality Sensitive Hashing for Beam Search on GPU},

   author={Shi, Xing and Xu, Shizhen and Knight, Kevin},

   year={2018},

   month={jun},

   archivePrefix={"arXiv"},

   primaryClass={cs.CL}

}

Download Download (PDF)   View View   Source Source   

1708

views

We present a GPU-based Locality Sensitive Hashing (LSH) algorithm to speed up beam search for sequence models. We utilize the winner-take-all (WTA) hash, which is based on relative ranking order of hidden dimensions and thus resilient to perturbations in numerical values. Our algorithm is designed by fully considering the underling architecture of CUDA-enabled GPUs (Algorithm/Architecture Co-design): 1) A parallel Cuckoo hash table is applied for LSH code lookup (guaranteed O(1) lookup time); 2) Candidate lists are shared across beams to maximize the parallelism; 3) Top frequent words are merged into candidate lists to improve performance. Experiments on 4 large-scale neural machine translation models demonstrate that our algorithm can achieve up to 4x speedup on softmax module, and 2x overall speedup without hurting BLEU on GPU.
Rating: 2.0/5. From 1 vote.
Please wait...

* * *

* * *

HGPU group © 2010-2024 hgpu.org

All rights belong to the respective authors

Contact us: