Semantic Scholar Open Access 2015 2317 sitasi

MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems

Tianqi Chen Mu Li Yutian Li Min Lin Naiyan Wang +5 lainnya

Abstrak

MXNet is a multi-language machine learning (ML) library to ease the development of ML algorithms, especially for deep neural networks. Embedded in the host language, it blends declarative symbolic expression with imperative tensor computation. It offers auto differentiation to derive gradients. MXNet is computation and memory efficient and runs on various heterogeneous systems, ranging from mobile devices to distributed GPU clusters. This paper describes both the API design and the system implementation of MXNet, and explains how embedding of both symbolic expression and tensor operation is handled in a unified fashion. Our preliminary experiments reveal promising results on large scale deep neural network applications using multiple GPU machines.

Topik & Kata Kunci

Penulis (10)

T

Tianqi Chen

M

Mu Li

Y

Yutian Li

M

Min Lin

N

Naiyan Wang

M

Minjie Wang

T

Tianjun Xiao

B

Bing Xu

C

Chiyuan Zhang

Z

Zheng Zhang

Format Sitasi

Chen, T., Li, M., Li, Y., Lin, M., Wang, N., Wang, M. et al. (2015). MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems. https://www.semanticscholar.org/paper/62df84d6a4d26f95e4714796c2337c9848cc13b5

Akses Cepat

PDF tidak tersedia langsung

Cek di sumber asli →
Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2015
Bahasa
en
Total Sitasi
2317×
Sumber Database
Semantic Scholar
Akses
Open Access ✓