Speeding up Convolutional Neural Network Training with Dynamic Precision Scaling and Flexible Multiplier-Accumulator

No abstract is available for this paper.

• Publication year

  2016

• Venue

  International Symposium on Low Power Electronics and Design

• Publication date

  2016-08-08

• Fields of study

  Computer Science, Engineering

• Identifiers
  DOI 10.1145/2934583.2934625
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Fixed Point Quantization of Deep Convolutional Networks

  2015influential reference
• Deep Learning with Limited Numerical Precision

  2015cited by this paper
• Neural Networks with Few Multiplications

  2015cited by this paper
• Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks

  2015cited by this paper
• cuDNN: Efficient Primitives for Deep Learning

  2014cited by this paper
• Caffe: Convolutional Architecture for Fast Feature Embedding

  2014influential reference
• Project Adam: Building an Efficient and Scalable Deep Learning Training System

  2014cited by this paper
• Training deep neural networks with low precision multiplications

  2014cited by this paper
• Minimizing Computation in Convolutional Neural Networks

  2014influential reference
• Going deeper with convolutions

  2014cited by this paper
• Large Scale Distributed Deep Networks

  2012cited by this paper
• ImageNet classification with deep convolutional neural networks

  2012cited by this paper
• Gradient-based learning applied to document recognition

  1998influential reference
• Dynamically scaled fixed point arithmetic

  1991influential reference
• Why systolic architectures?

  1982cited by this paper

• A Survey and Comparative Analysis of Number Systems for Deep Neural Networks

  2025influential citation
• Resource constrained neural network training

  2024cites this paper
• Adaptive approximate computing in edge AI and IoT applications: A review

  2024cites this paper
• Error Resilience in Deep Neural Networks Using Neuron Gradient Statistics

  2024cites this paper
• Number Systems for Deep Neural Network Architectures: A Survey

  2023influential citation
• Neural network training with limited precision and asymmetric exponent

  2022cites this paper
• Based on DeepLab v3+ model to realize the road scene semantic segmentation

  2022cites this paper
• HECATE: Performance-Aware Scale Optimization for Homomorphic Encryption Compiler

  2022cites this paper
• Hartley Stochastic Computing For Convolutional Neural Networks

  2021cites this paper
• Improving the Accuracy and Hardware Efficiency of Neural Networks Using Approximate Multipliers

  2020cites this paper
• AccPar: Tensor Partitioning for Heterogeneous Deep Learning Accelerators

  2020cites this paper
• A Novel Signed Bit-serial Fixed-point Accumulator with Configurable Overflow-Protection Precision

  2019cites this paper
• Design and Analysis of a Neural Network Inference Engine Based on Adaptive Weight Compression

  2019cites this paper
• HyPar: Towards Hybrid Parallelism for Deep Learning Accelerator Array

  2019cites this paper
• Design of Reliable DNN Accelerator with Un-reliable ReRAM

  2019influential citation
• Enabling Energy-Efficient Unsupervised Monocular Depth Estimation on ARMv7-Based Platforms

  2019cites this paper
• Research on the Current Status of Sparse Neural Network Acceleration Processing Technology in Deep Learning

  2019cites this paper
• Dynamic Precision Tunability in Low-Power Training of Deep Neural Networks

  2019cites this paper
• Energy efficient, secure and noise robust deep learning for the internet of things

  2018influential citation
• Investigating the Effects of Dynamic Precision Scaling on Neural Network Training

  2018influential citation
• Quantization Error as a Metric for Dynamic Precision Scaling in Neural Net Training

  2018influential citation
• Deploying Customized Data Representation and Approximate Computing in Machine Learning Applications

  2018cites this paper
• Precision Scaling of Neural Networks for Efficient Audio Processing

  2017cites this paper
• Convolutional Neural Network for Freight Train Information Recognition

  2017cites this paper
• Design of an energy-efficient accelerator for training of convolutional neural networks using frequency-domain computation

  2017cites this paper
• On-chip training of recurrent neural networks with limited numerical precision

  2017cites this paper
• of Systems Architecture

  year unknowncites this paper