Sparse Persistent RNNs: Squeezing Large Recurrent Networks On-ChipDownload PDF

15 Feb 2018 (modified: 23 Feb 2018)ICLR 2018 Conference Blind SubmissionReaders: Everyone
Abstract: Recurrent Neural Networks (RNNs) are powerful tools for solving sequence-based problems, but their efficacy and execution time are dependent on the size of the network. Following recent work in simplifying these networks with model pruning and a novel mapping of work onto GPUs, we design an efficient implementation for sparse RNNs. We investigate several optimizations and tradeoffs: Lamport timestamps, wide memory loads, and a bank-aware weight layout. With these optimizations, we achieve speedups of over 6x over the next best algorithm for a hidden layer of size 2304, batch size of 4, and a density of 30%. Further, our technique allows for models of over 5x the size to fit on a GPU for a speedup of 2x, enabling larger networks to help advance the state-of-the-art. We perform case studies on NMT and speech recognition tasks in the appendix, accelerating their recurrent layers by up to 3x.
TL;DR: Combining network pruning and persistent kernels into a practical, fast, and accurate network implementation.
Keywords: Sparsity, Pruning, Compression, RNN, LSTM, Persistent, RF-Resident, GPU
9 Replies