Non-Volatile Memory Array Based Quantization- and Noise-Resilient LSTM Neural Networks

02/25/2020
by   Wen Ma, et al.
0

In cloud and edge computing models, it is important that compute devices at the edge be as power efficient as possible. Long short-term memory (LSTM) neural networks have been widely used for natural language processing, time series prediction and many other sequential data tasks. Thus, for these applications there is increasing need for low-power accelerators for LSTM model inference at the edge. In order to reduce power dissipation due to data transfers within inference devices, there has been significant interest in accelerating vector-matrix multiplication (VMM) operations using non-volatile memory (NVM) weight arrays. In NVM array-based hardware, reduced bit-widths also significantly increases the power efficiency. In this paper, we focus on the application of quantization-aware training algorithm to LSTM models, and the benefits these models bring in terms of resilience against both quantization error and analog device noise. We have shown that only 4-bit NVM weights and 4-bit ADC/DACs are needed to produce equivalent LSTM network performance as floating-point baseline. Reasonable levels of ADC quantization noise and weight noise can be naturally tolerated within our NVMbased quantized LSTM network. Benchmark analysis of our proposed LSTM accelerator for inference has shown at least 2.4x better computing efficiency and 40x higher area efficiency than traditional digital approaches (GPU, FPGA, and ASIC). Some other novel approaches based on NVM promise to deliver higher computing efficiency (up to 4.7x) but require larger arrays with potential higher error rates.

READ FULL TEXT

page 1

page 3

research
02/14/2022

Vau da muntanialas: Energy-efficient multi-die scalable acceleration of RNN inference

Recurrent neural networks such as Long Short-Term Memories (LSTMs) learn...
research
04/02/2020

Device-aware inference operations in SONOS nonvolatile memory arrays

Non-volatile memory arrays can deploy pre-trained neural network models ...
research
06/15/2021

Exploring the Feasibility of Using 3D XPoint as an In-Memory Computing Accelerator

This paper describes how 3D XPoint memory arrays can be used as in-memor...
research
02/27/2018

A High GOPs/Slice Time Series Classifier for Portable and Embedded Biomedical Applications

Modern wearable rehabilitation devices and health support systems operat...
research
03/09/2020

Software-Level Accuracy Using Stochastic Computing With Charge-Trap-Flash Based Weight Matrix

The in-memory computing paradigm with emerging memory devices has been r...
research
01/07/2021

BRDS: An FPGA-based LSTM Accelerator with Row-Balanced Dual-Ratio Sparsification

In this paper, first, a hardware-friendly pruning algorithm for reducing...
research
10/19/2021

PR-CIM: a Variation-Aware Binary-Neural-Network Framework for Process-Resilient Computation-in-memory

Binary neural networks (BNNs) that use 1-bit weights and activations hav...

Please sign up or login with your details

Forgot password? Click here to reset