Search Results for author:
Found 18 papers, 12 papers with code
Optimal Clipping and Magnitude-aware Differentiation for Improved Quantization-aware Training
Data clipping is crucial in reducing noise in quantization operations and improving the achievable accuracy of quantization-aware training (QAT).
PatchNet -- Short-range Template Matching for Efficient Video Processing
Object recognition is a fundamental problem in many video processing tasks, accurately locating seen objects at low computation cost paves the way for on-device video recognition.
VS-Quant: Per-vector Scaled Quantization for Accurate Low-Precision Neural Network Inference
4-bit weights and 8-bit activations achieve near-full-precision accuracy for both BERT-base and BERT-large on SQuAD while reducing area by 26% compared to an 8-bit baseline.
SpArch: Efficient Architecture for Sparse Matrix Multiplication
We then propose a condensed matrix representation that reduces the number of partial matrices by three orders of magnitude and thus reduces DRAM access by 5. 4x.
Hardware Architecture Distributed, Parallel, and Cluster Computing
A Delay Metric for Video Object Detection: What Average Precision Fails to Tell
Average precision (AP) is a widely used metric to evaluate detection accuracy of image and video object detectors.
Analog/Mixed-Signal Hardware Error Modeling for Deep Learning Inference
We also introduce an energy model to predict the requirements of high-accuracy AMS hardware running large networks and use it to show that for ADC-dominated designs, there is a direct tradeoff between energy efficiency and network accuracy.
CaTDet: Cascaded Tracked Detector for Efficient Object Detection from Video
Experiments on the KITTI dataset show that CaTDet reduces operation count by 5. 1-8. 7x with the same mean Average Precision(mAP) as the single-model Faster R-CNN detector and incurs additional delay of 0. 3 frame.
Efficient Sparse-Winograd Convolutional Neural Networks
First, we move the ReLU operation into the Winograd domain to increase the sparsity of the transformed activations.
Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
The situation gets even worse with distributed training on mobile devices (federated learning), which suffers from higher latency, lower throughput, and intermittent poor connections.
Exploring the Regularity of Sparse Structure in Convolutional Neural Networks
Since memory reference is more than two orders of magnitude more expensive than arithmetic operations, the regularity of sparse structure leads to more efficient hardware design.
SCNN: An Accelerator for Compressed-sparse Convolutional Neural Networks
Convolutional Neural Networks (CNNs) have emerged as a fundamental technology for machine learning.
Trained Ternary Quantization
To solve this problem, we propose Trained Ternary Quantization (TTQ), a method that can reduce the precision of weights in neural networks to ternary values.
ESE: Efficient Speech Recognition Engine with Sparse LSTM on FPGA
Evaluated on the LSTM for speech recognition benchmark, ESE is 43x and 3x faster than Core i7 5930k CPU and Pascal Titan X GPU implementations.
DSD: Dense-Sparse-Dense Training for Deep Neural Networks
We propose DSD, a dense-sparse-dense training flow, for regularizing deep neural networks and achieving better optimization performance.
SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
(2) Smaller DNNs require less bandwidth to export a new model from the cloud to an autonomous car.
Ranked #1 on
Image Classification
on ImageNet-P
EIE: Efficient Inference Engine on Compressed Deep Neural Network
EIE has a processing power of 102GOPS/s working directly on a compressed network, corresponding to 3TOPS/s on an uncompressed network, and processes FC layers of AlexNet at 1. 88x10^4 frames/sec with a power dissipation of only 600mW.
Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding
To address this limitation, we introduce "deep compression", a three stage pipeline: pruning, trained quantization and Huffman coding, that work together to reduce the storage requirement of neural networks by 35x to 49x without affecting their accuracy.
Learning both Weights and Connections for Efficient Neural Networks
On the ImageNet dataset, our method reduced the number of parameters of AlexNet by a factor of 9x, from 61 million to 6. 7 million, without incurring accuracy loss.
