Success: Thank you for subscribing!
You will begin receiving emails as new content is posted. You may unsubscribe any time by clicking the link in the email.

Kubeflow news

Your weekly update of curated kubeflow news from across the web.

Contact us with your kubeflow news

Digest #2019.12.16 –The Decade of Artifical Intelligence

16 December 2019

AI Competing with Wall Street –  Artificial Intelligence is replacing traders on Wall Street; what used to be a busy bustling market with chatter and cold calls are now the slight hums of machines running algorithms and executing trades. The demand for people with coding skills who can train models for intelligent trading and fight fraud and security risks is on the rise. Investment banks that once called for traders and placed big bets on them are now pivoting to place their bets on people who have the math, technology, software, coding, data analytics and related skills to work along with electronic trading.

Artificial Intelligence Decade Review – 2010s was a big decade for AI, ML and robots. There were big breakthroughs, tech got cheaper, more convenient, at times more confusing. We made the machines smarter, more powerful, invited them in our homes, put them in control of minor everyday tasks, they also leaked some data at times and then we tried to control them (GDPR…). Read the full article for the decade in review for Artificial Intelligence and Machine Learning.

Reducing Maintainer Toil on Kubeflow – Maintaining Open-Source Software is hard! Kubeflow’s Github repo is seeing an exponentially increasing number of open issues and requests as it is gaining traction. To ease maintainability a bit, the team at Github and Kubeflow OSS introduced Project code-intelligence, the idea for which started out of demos for Kubeflow use-cases. If you’re interested in helping Kubeflow OSS “Help us create reports that pull and visualize key performance indicators (KPI), Github link. Ensemble repo specific and non-repo specific label predictions, Github link.

Everyday Applications of Machine Learning – Machine Learning and Artificial Intelligence have become a big part of most people’s everyday lives. From the searches on the internet to the targeted ads and your vacuum cleaner mapping out your room, AI/ML in technology is a prevailing part of our daily lives. Check out 12 applications of AI/ML that are part of some daily lives and business.



Digest #2019.12.02 – Machine Learning VS Minecraft

2 December 2019

Google pushing TPU with new TensorFlow release – TensorFlow 2.1 is here, welcome to the TPU push; that’s good news for many. We’ll also see improved performance on Linux and Windows and GPU support out-the-box. There are a few changes and issues to be aware of and if you’re using Python 2, TF 2.1 will be the last release you can work with. Check out the summary of changes here.

Machine Learning = Artificial Intelligence? – The article is an interesting read discussing whether machine learning = artificial intelligence. Machine Learning and Artificial Intelligence have a lot of overlap amongst them but they’re not exactly the same. MIT Professor Luis Perez-Breva argues that , “most of what is currently being branded as AI in the market and media is not AI at all, but rather just different versions of ML where the systems are being trained to do a specific, narrow task, using different approaches to ML, of which Deep Learning is currently the most popular.”

Machine Learning Books to Read– For all of us reading fans and bibliophiles interested in Machine Learning, check out this fancy list of machine learning books to read. We’ve personally read most of them and would highly recommend if you’re interested in exploring you ML chops a bit further.

Machine Learning takes on Minecraft – Minecraft vs Machine Learning, fun times! Machine Learning took on the popular game of Minecraft in an intensive gaming competition. The competition allowed coders to devise imitation learning strategies that will help machines win over human players. Imitation learning is the form of machine learning also found in autonomous driving. Check out the full article for an interesting read.

Robots Debating About AI – Check out robots making arguments for and against the dangers of Artificial Intelligence. The demonstration is by Project Debater, an IBM robot designed to hold arguments based on human input. The robots debated using arguments based on 1100 human submissions. The pro-AI side narrowly won the argument by 51.22% audience votes, fascinating!


Kubeflow Talks at Kubecon San Diego 1/2

25 November 2019

Throwback to Kubecon last week where Kubeflow was the most talked about topic other than Kubernetes itself. It was great to see so much excitement around Kubeflow. For those of you who missed the event, or were too busy to catch these talks, or just couldn’t be everywhere exciting at once, below is a list of are recordings of Kubeflow talks from last week in San Diego.

Building and Managing a Centralized Kubeflow Platform at Spotify – Keshi Dai & Ryan Clough, Spotify

Panel: Enterprise-grade, On-prem Kubeflow in the Financial Sector

Kubeflow: Multi-Tenant, Self-Serve, Accelerated Platform for Practitioners- Kam Kasravi & Kunming Qu

Building a Medical AI with Kubernetes and Kubeflow – Jeremie Vallee, Babylon Health

Supercharge Kubeflow Performance on GPU Clusters – Meenakshi Kaushik & Neelima Mukiri, Cisco



Kubeflow Talks at Kubecon San Diego 2/2

25 November 2019

KubeFlow’s Serverless Component: 10x Faster, a 1/10 of the Effort – Orit Nissan-Messing, Iguazio

Enabling Kubeflow with Enterprise-Grade Auth for On-Prem – Yannis Zarkadas & Krishna Durai

Measuring and Optimizing Kubeflow Clusters at Lyft – Konstantin Gizdarski & Richard Liu

Towards Continuous Computer Vision Model Improvement with Kubeflow – Derek Hao Hu & Yanjia Li

Tutorial: From Notebook to Kubeflow Pipelines – Jeremy Lewi, Michelle Casbon, Stefano Fioravanzo & Ilias Katsakioris

Advanced Model Inferencing Leveraging KNative, Istio & Kubeflow Serving – Animesh Singh & Clive Cox



Digest #2019.10.28 – AI for Biodiversity Research

28 October 2019

AI for Biodiversity Research – Google, in collaboration with Global Biodiversity Information Facility (GBIF), iNaturalist, and Visipedia is making a push to bring AI to biodiversity research. While ML is prevalent in biodiversity research, proper attribution and oversight is a hit or miss. Google is hoping to bridge the gap and raise the academic bar. With the growing importance and awareness of fairness,ethics, and transparency, the innovative uses of machine learning for biodiversity stands to be a challenging, exciting and rewarding task.

A Beginners’ Guide to Machine Learning– This article serves as an interesting machine learning 101. The article breaks down machine learning into three types; classification, regression and unsupervised learning with examples in Python using the library scikit-learn. While there is a lot more to delve into, this is a quick intro to the ML world, or a nice refresher if you have been away for a while.

Polynote by Netflix – Improve notebook execution, code quality in an IDE environment, meet Netflix Polynote! Open-source software that empowers data science and machine learning; super exciting experiments lie ahead. “Plenty of exciting work lies ahead, we are very optimistic about the potential of Polynote, and we hope to learn from the community just as much as we hope they will find value from Polynote.” Check it out on their Github.

AI and Health – The article sheds light on an interesting perspective with the use of AI to improve life spans. Imagine the possibility of knowing our risk profiles when we are born, the types of diseases we could be prone to and preventive measures taken years in advance. However, this will be no easy feat by any means, it will require social inclusiveness, strict governance of AI and data and diversity across the spectrum of devices, code, and human forms. Nonetheless, the idea itself and the nascent research is an exciting take on how AI can improve our lives and make the world a better place.