Cited 0 time in
Cited 4 time in
Peregrine: A flexible hardware accelerator for LSTM with limited synaptic connection patterns
- Peregrine: A flexible hardware accelerator for LSTM with limited synaptic connection patterns
- Kung, Jaeha; Park, Junki; Park, Sehun; Kim, Jae-Joon
- DGIST Authors
- Kung, Jaeha
- Issue Date
- Design Automation Conference
- In this paper, we present an integrated solution to design a highperformance LSTM accelerator.We propose a fast and flexible hardware architecture, named Peregrine, supported by a stack of innovations from algorithm to hardware design. Peregrine first minimizes the memory footprint by limiting the synaptic connection patterns within the LSTM network. Also, Peregrine provides parallel Huffman decoders with adaptive clocking to provide flexibility in dealing with a wide range of sparsity levels in the weight matrices. All these features are incorporated in a novel hardware architecture to maximize energy-efficiency. As a result, Peregrine improves performance by ~38% and energy-efficiency by ~33% in speech recognition compared to the state-of-the-art LSTM accelerator. © 2019 Association for Computing Machinery.
- Institute of Electrical and Electronics Engineers Inc.
- Related Researcher
Intelligent Digital Systems Lab
딥러닝, 가속하드웨어, 저전력 하드웨어, 고성능 시스템
There are no files associated with this item.
- Department of Information and Communication EngineeringIntelligent Digital Systems Lab2. Conference Papers
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.