Loading…

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

AI/ML/DL [clear filter]
Tuesday, October 27
 

18:30 GMT

Using Volcano and Kubernetes for Cutting-Edge AI Deployment - Yedong Liu & William Wang, Huawei
Over the past few year, cloud native software brought many benefits to industry for deployment, management etc. MindSpore, a new open source deep learning framework, is also one of them. As it suggests, MindSpore is working on collaboration with CNCF projects such as Kubernetes and Volcano to allow deploying MindSpore job on container environment. This session will show the technical details as well as the examples on how a MindSpore training job runs in the container env and distributed GPU demo with Volcano. Together with long term goals of MindSpore & other cloud native projects are also included in this session.

Speakers
YL

Yedong Liu

Open source engineer, Huawei
Yedong Liu is an Open Source Engineer from Huawei, he participated in Open Source communities including ONNX, Volcano etc. He is now a member of the MindSpore community which is a newly open sourced deep learning framework. Yedong is working on bringing more convenience to the developers... Read More →
WW

William Wang

Architect, Huawei
Volcano member, software developer, currently working on Volcano Job scheduling across multiple cluster development. Experienced in batch system and Bigdata, AI worklaod performance acceleration. Spearker in ArchiSumit sharing spark on kubernetes best practice.


Tuesday October 27, 2020 18:30 - 19:20 GMT
AI/ML/DL Theater
  AI/ML/DL, Machine and Deep Learning
  • Skill Level Any
  • Technical Talk No
 
Wednesday, October 28
 

19:30 GMT

How Jina Saves Your Time on Building Cloud-Native Neural Search Systems - Han Xiao, Jina AI
Today, with the ever more long documents and multimedia data, finding the right information is more important and challenging than ever. The rise of deep learning has ushered in a new era of "neural search". However, building a neural search system is non-trivial work for most of the engineers. The main challenges are: (1) long dev cycle due to the complex tech stack (2) poor scalability due to the glued-architecture (3) strong requirements on the domain knowledge to fine-tune the results. With Jina (https://github.com/jina-ai/jina), engineers can quickly build up a search engine powered by state-of-the-art AI in just minutes. In this talk, I will introduce the design philosophy and the key features of Jina; and showcase how Jina bootstraps a QA semantic search system and a short-video search system in just lines of code.

Speakers
HX

Han Xiao

CEO, Jina AI
Dr. Han Xiao is the Founder & CEO of Jina AI. Han has worked in AI OSS for quite some time. His Fashion-MNIST and bert-as-service were listed as the most popular AI open-source projects in 2017&18 world-widely. In 2018-2020, Han led a team on neural information retrieval at Tencent... Read More →


Wednesday October 28, 2020 19:30 - 20:20 GMT
AI/ML/DL Theater
  AI/ML/DL, Machine and Deep Learning
 

Twitter Feed