Energy-Efficient ConvNets Through Approximate Computing

2016-03-22 · Bert Moons, Bert de Brabandere, Luc van Gool, Marian Verhelst

Recently ConvNets or convolutional neural networks (CNN) have come up as state-of-the-art classification and detection algorithms, achieving near-human performance in visual detection. However, ConvNet algorithms are typically very computation and memory intensive. In order to be able to embed ConvNet-based classification into wearable platforms and embedded systems such as smartphones or ubiquitous electronics for the internet-of-things, their energy consumption should be reduced drastically. This paper proposes methods based on approximate computing to reduce energy consumption in state-of-the-art ConvNet accelerators. By combining techniques both at the system- and circuit level, we can gain energy in the systems arithmetic: up to 30x without losing classification accuracy and more than 100x at 99% classification accuracy, compared to the commonly used 16-bit fixed point number format.

📄 PDF Abstract BibTeX arXiv:1603.06777

Code (0)

등록된 구현이 없습니다.

Tasks

Classification General Classification

Deep Symmetry Networks

2014-12-01 · NeurIPS 2014 12 · Robert Gens, Pedro M. Domingos

The chief difficulty in object recognition is that objects' classes are obscured by a large number of extraneous sources of variability, such as pose and part deformation. These sources of variation can be represented by…

Object Recognition Translation

Quality Scalable Quantization Methodology for Deep Learning on Edge

2024-07-15 · Salman Abdul Khaliq, Rehan Hafiz

Deep Learning Architectures employ heavy computations and bulk of the computational energy is taken up by the convolution operations in the Convolutional Neural Networks. The objective of our proposed work is to reduce t…

Deep Learning Edge-computing Quantization

Instabilities in Convnets for Raw Audio

2023-09-11 · Daniel Haider, Vincent Lostanlen, Martin Ehler, Peter Balazs

What makes waveform-based deep learning so hard? Despite numerous attempts at training convolutional neural networks (convnets) for filterbank design, they often fail to outperform hand-crafted baselines. These baselines…

Audio Signal Processing

Exploiting In-Sensor Computing for Energy-Efficient Earth Observation

2026-05-31 · Luigi Capogrosso, Pietro Bonazzi, Loris Hoxhaj, Michele Magno arxiv

The rapid growth of the satellite industry has driven a significant increase in geospatial data acquisition, highlighting a critical bottleneck: the severe disparity between the volume of collected sensor data and the li…

ADIC: Anomaly Detection Integrated Circuit in 65nm CMOS utilizing Approximate Computing

2020-08-21 · Bapi Kar, Pradeep Kumar Gopalakrishnan, Sumon Kumar Bose, Mohendra Roy 외

In this paper, we present a low-power anomaly detection integrated circuit (ADIC) based on a one-class classifier (OCC) neural network. The ADIC achieves low-power operation through a combination of (a) careful choice of…

Anomaly Detection One-class classifier