- 1、本文档共9页,可阅读全部内容。
- 2、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。
- 3、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载。
- 4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
查看更多
MarianaTencentDeepLearningPlatformanditsApplicationsMarianaTencentDeepLearningPlatformanditsApplications
Mariana: Tencent Deep Learning Platform and its ApplicationsABSTRACT Deep learning gains lots of attentions in recent years and is more and more important for mining values in big data. However, to make deep learning practical for a wide range of applications in Tencent Inc., three requirements must be considered: 1) Lots of computational power are required to train a practical model with tens of millions of parameters and billions of samples for products such as automatic speech recognition (ASR), and the number of parameters and training data is still growing. 2) The capability of training larger model is necessary for better model quality. 3) Easy to use frameworks are valuable to do many experiments to perform model selection, such as finding an appropriate optimization algorithm and tuning optimal hyper-parameters. To accelerate training, support large models, and make experiments easier, we built Mariana, the Tencent deep learning platform, which utilizes GPU and CPU cluster to train models parallelly with three frameworks: 1) a multi-GPU data parallelism framework for deep neural networks (DNNs). 2) a multi-GPU model parallelism and data parallelism framework for deep convolutional neural networks (CNNs). 3) a CPU cluster framework for large scale DNNs. Mariana also provides built-in algorithms and features to facilitate experiments. Mariana is in production usage for more than one year, achieves state-of-the-art acceleration performance, and plays a key role in training models and improving quality for automatic speech recognition and image recognition in TencentWeChat, a mobile social platform, and for Ad click-through rate prediction (pCTR) in Tencent QQ, an instant messaging platform, and TencentQzone, a social networking service.1. INTRODUCTION 1.1 BackgroudTencent provides a wide range of Internet services, such as WeChat, a mobile social platform having 396 millions of monthly active users (MAU), QQ, an instant messaging platform having 848
您可能关注的文档
- Lecture Notes S2 - 4.doc
- Lecture Notes Shelley 1.doc
- lecture notes Module 4 part1.doc
- Lecture Notes S2 - 8.doc
- Lecture18.doc
- Lecture1presentationsmall.doc
- Lecture Notes S2 - 9.doc
- lecture 物理类话题.doc
- Lecture4 Imagery.doc
- Lecture Notes_U5.doc
- 2024年江西省寻乌县九上数学开学复习检测模拟试题【含答案】.doc
- 2024年江西省省宜春市袁州区数学九上开学学业水平测试模拟试题【含答案】.doc
- 《GB/T 44275.2-2024工业自动化系统与集成 开放技术字典及其在主数据中的应用 第2部分:术语》.pdf
- 中国国家标准 GB/T 44275.2-2024工业自动化系统与集成 开放技术字典及其在主数据中的应用 第2部分:术语.pdf
- GB/T 44285.1-2024卡及身份识别安全设备 通过移动设备进行身份管理的构件 第1部分:移动电子身份系统的通用系统架构.pdf
- 《GB/T 44285.1-2024卡及身份识别安全设备 通过移动设备进行身份管理的构件 第1部分:移动电子身份系统的通用系统架构》.pdf
- 中国国家标准 GB/T 44285.1-2024卡及身份识别安全设备 通过移动设备进行身份管理的构件 第1部分:移动电子身份系统的通用系统架构.pdf
- GB/T 44275.11-2024工业自动化系统与集成 开放技术字典及其在主数据中的应用 第11部分:术语制定指南.pdf
- 中国国家标准 GB/T 44275.11-2024工业自动化系统与集成 开放技术字典及其在主数据中的应用 第11部分:术语制定指南.pdf
- 《GB/T 44275.11-2024工业自动化系统与集成 开放技术字典及其在主数据中的应用 第11部分:术语制定指南》.pdf
文档评论(0)