Google Cloud Platform Podcast

Nenad Stojanovski and Andrew Hoying join Mark and Melanie

Bryan Catanzaro, the VP Applied Deep Learning Research at NVIDIA, joins Mark and Melanie this week to discuss how his team uses applied deep learning to make NVIDIA products and processes better. We talk about parallel processing and compute with GPUs as well as his team's research in graphics, text and audio to change how these forms of communication are created and rendered by using deep learning.

This week we are also joined by a special co-host, Sherol Chen who is a developer advocate on GCP and machine learning researcher on Magenta at Google. Listen at the end of the podcast where Mark and Sherol chat about all things GDC.

Bryan Catanzaro is VP of Applied Deep Learning Research at NVIDIA, where he leads a team solving problems in domains ranging from video games to chip design using deep learning. Bryan earned his PhD from Berkeley, where he focused on parallel computing, machine learning, and programming models. He earned his MS and BS from Brigham Young University, where he worked on higher radix floating-point representations for FPGAs. Bryan worked at Baidu to create next generation systems for training and deploying deep learning models for speech recognition. Before that, he was a researcher at NVIDIA, where he worked on programming models for parallel processors, as well as libraries for deep learning, which culminated in the creation of the widely used CUDNN library.

• NVIDIA Tesla V100s coming to Google Cloud site
• Automatic Serverless Deployment with Cloud Source Repositories blog
• Magenta site
  • NSynth Super site
  • MusicVAE site
  • Making music using new sounds generated with machine learnnig blog
• Building Blocks of Interpretability blog

• NVIDIA site
• NVIDIA GPU Technology Conference (GTC) site
• CUDA site
• cuDNN site
• NVIDIA Volta site
• NVIDIA Tesla P4 docs
• NVIDIA Tesla V100s site
• Silicon Valley AI Lab Baidu Research site
• ICML: International Conference on Machine Learning site
• CVPR: Computer Vision and Pattern Recognition Conference site

Referenced Papers & Research:

• Deep learning with COTS HPC System paper
• Building High-level Features Using Large Scale Unsupervised Learning paper
• OpenAI Learning to Generate Reviews and Discovering Sentiment paper
• Progressive Growing of GANs for Improved Quality, Stability, and Variation paper and CelebA dataset
• High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs paper
• Deep Image Prior site
• How a Japanese cucumber farmer is using deep learning and TensorFlow blog

Sample Talks:

• Future of AI Hardware Panel video
• High Performance Computing is Supercharging AI blog/video
• AI Podcast: Where is Deep Learning Going Next? blog/video

Sample Resources:

• Coursera How Google does Machine Learning site
• NVIDIA Deep Learning Institute site
• Udacity AI Nanodegree site
• Kaggle site
• TensorFlow site
• PyTorch site
• Keras site

What to watch out for and get involved in at the Game Developers Conference (GDC) this year and in the future?

• International Grame Developers Association (IGDA) site
• Fellowship of GDC Parties site
• ALtCtrlGDC site
• Experimental Gameplay Workshop site
• Women in Games International (WIGI) site
• Blacks in Gaming (BIG) site
• Serious Games (SIGs) site
• What's New in Firebase and Google Cloud Platform for Games site
• Summits to Checkout:
  • AI Game Developers Summit site
  • Game Narrative Summit site
  • Independent Games Summit site
• Additional Advice:
  • The first two days are summits which are great because topic focused
  • Expo floor takes a good hour to get through
  • WIGI, BIG and SIGs (Google and Microsoft) have the best food
  • GDC is composed of various communities
  • Bring business cards
  • Check out post-mortems
• Favorite Games:
  • Mass Effect site
  • Final Fantasy wiki
• Games Mark & Sherol are currently playing:
  • Hearthstone site
  • Dragon Age Origins wiki

Mark and Sherol are at the Game Developer's Conference (GDC). You can find them via the Google at GDC 2018 site.

Sherol will be at TensorFlow Dev Summit speaking about machine learning research and creativity next week.

Bryan Catanzaro, the VP Applied Deep Learning Research at NVIDIA, joins Mark and Melanie this week to discuss how his team uses applied deep learning to make NVIDIA products and processes better. We talk about parallel processing and compute with GPUs as well as his team’s research in graphics, text and audio to change how these forms of communication are created and rendered by using deep learning.

This week we are also joined by a special co-host, Sherol Chen who is a developer advocate on GCP and machine learning researcher on Magenta at Google. Listen at the end of the podcast where Mark and Sherol chat about all things GDC.

Product Manager Morgan McLean and Software Engineer JBD join Melanie and Mark this week to discuss the new open source project OpenCensus, a single distribution of libraries for metrics and distributed tracing with minimal overhead that allows you to export data to multiple backends.

Morgan McLean is the Product Manager for Tracing, Debugging, and Profiling at Google, including OpenCensus

JBD leads the OpenCensus Go and our integrations with other projects.

• Introducing Agones: Open-source, multiplayer, dedicated game-server hosting built on Kubernetes blog github
• TensorFlow 1.6 release summit
• TensorFlow Object Detection on iOS youtube
• Optimizing your Cloud Storage performance: Google Cloud Performance Atlas blog
• Introducing GCP's new interactive CLI blog docs

• OpenCensus site docs
• Github Organisation github
• How Google uses Census internally blog
• The value of OpenCensus blog
• Tracing
  • Zipkin site
  • Stackdriver site docs
  • jaeger github
• Metrics
  • Stackdriver docs
  • Prometheus site
• Google Summer of Code site

OpenCensus

I heard there are abilities to natively extend Kubernetes - what does that mean, and also how do I do that?

• Custom Resource Definitions reference guide
• Sample Controller github
• TGIK Custom Controller video series github youtube youtube youtube
• An Introduction to Extending Kubernetes with CustomResourceDefinitions blog
• memcached-operator
• etcd-operator

Mark will be at the Game Developer's Conference | GDC in March. You can find him via the Google at GDC 2108 site.

Product Manager Morgan McLean and Software Engineer JBD join Melanie and Mark this week

Dr. Fei-Fei Li, the Chief Scientist of AI/ML at Google joins Melanie and Mark this week to talk about how Google enables businesses to solve critical problems through AI solutions. We talk about the work she is doing at Google to help reduce AI barriers to entry for enterprise, her research with Stanford combining AI and health care, where AI research is going, and her efforts to overcome one of the key challenges in AI by driving for more diversity in the field.

Dr. Fei-Fei Li is the Chief Scientist of AI/ML at Google Cloud. She is also an Associate Professor in the Computer Science Department at Stanford, and the Director of the Stanford Artificial Intelligence Lab. Dr. Fei-Fei Li's main research areas are in machine learning, deep learning, computer vision and cognitive and computational neuroscience. She has published more than 150 scientific articles in top-tier journals and conferences, including Nature, PNAS, Journal of Neuroscience, CVPR, ICCV, NIPS, ECCV, IJCV, IEEE-PAMI, etc. Dr. Fei-Fei Li obtained her B.A. degree in physics from Princeton in 1999 with High Honors, and her PhD degree in electrical engineering from California Institute of Technology (Caltech) in 2005. She joined Stanford in 2009 as an assistant professor, and was promoted to associate professor with tenure in 2012. Prior to that, she was on faculty at Princeton University (2007-2009) and University of Illinois Urbana-Champaign (2005-2006). Dr. Li is the inventor of ImageNet and the ImageNet Challenge, a critical large-scale dataset and benchmarking effort that has contributed to the latest developments in deep learning and AI. In addition to her technical contributions, she is a national leading voice for advocating diversity in STEM and AI. She is co-founder of Stanford's renowned SAILORS outreach program for high school girls and the national non-profit AI4ALL. For her work in AI, Dr. Li is a speaker at the TED2015 main conference, a recipient of the IAPR 2016 J.K. Aggarwal Prize, the 2016 nVidia Pioneer in AI Award, 2014 IBM Faculty Fellow Award, 2011 Alfred Sloan Faculty Award, 2012 Yahoo Labs FREP award, 2009 NSF CAREER award, the 2006 Microsoft Research New Faculty Fellowship and a number of Google Research awards. Work from Dr. Li's lab have been featured in a variety of popular press magazines and newspapers including New York Times, Wall Street Journal, Fortune Magazine, Science, Wired Magazine, MIT Technology Review, Financial Times, and more. She was selected as a 2017 Women in Tech by the ELLE Magazine, a 2017 Awesome Women Award by Good Housekeeping, a Global Thinker of 2015 by Foreign Policy, and one of the "Great Immigrants: The Pride of America" in 2016 by the Carnegie Foundation, past winners include Albert Einstein, Yoyo Ma, Sergey Brin, et al.

• Terah Lyons appointed founding executive director of Partnership on AI article & site
• Fully managed export and import with Cloud Datastore now generally available blog
• How Color uses the new Variant Transforms tool for breakthrough clinical data science with BigQuery blog & repo
• Google Cloud and NCAA team up for a unique March Madness copmetition hosted on Kaggle blog

• AI4All site, they are hiring and how to become a mentor
• Cloud AI site
• Cloud AutoML site
• Cloud Vision API site and docs
• Cloud Speech API site and docs
• Cloud Natural Language API site and docs
• Cloud Translation API site and docs
• Cloud Machine Learning Engine docs
• TensorFlow site, github and Dev Summit waitlist
• ImageNet site & Kaggle ImageNet Competition site
• Stanford Medicine site & Stanford Children's Hospital site

Additional sample resources on Dr. Fei-Fei Li:

• Citations site
• Stanford Vision Lab site
• Fei-Fei Li | 2018 MAKERS Conference video
• Google Cloud's Li Sees Transformative Time for Enterprise video
• Past, Present and Future of AI / Machine Learning Google I/O video
• Research Symposium 2017 - Morning Keynote Address at Harker School video
• How we're teaching computers to understand pictures video
• Melinda Gates and Fei-Fei Li Want to Liberate AI from "Guy with Hoodies" article

Dr. Fei-Fei Li

Where can I learn more about machine learning?

Listing of some of the many resources out there in no particular order:

• How Google does Machine Learning coursera
• Machine Learning with Andrew Ng coursera and Deep Learning Specialization coursera
• fast.ai site
• Machine Learning with John W. Paisley edx
• Machine Learning Columbia University edx

International Women's Day site covers information on events in your area, and additional resources.

Sample of recent women in tech events to keep on radar for next year:

• Women Techmakers site
• Lesbians Who Tech site
• Women in Data Science Conference site

Mark will be at the Game Developer's Conference | GDC in March.

Dr. Fei-Fei Li, the Chief Scientist of AI/ML at Google joins Melanie and Mark this week to talk about how Google enables businesses to solve critical problems through AI solutions. We talk about the work she is doing at Google to help reduce AI barriers to entry for enterprise, her research with Stanford combining AI and health care, where AI research is going, and her efforts to overcome one of the key challenges in AI by driving for more diversity in the field.

We have the pleasure this week of having the Director of Solutions for Google Cloud Miles Ward and Cloud Solutions Architect Grace Mollison join Mark and Melanie to discuss Solution Architects, what they do and how they interact with Customers at Google Cloud Platform.

Miles Ward is a three-time technology startup entrepreneur with a decade of experience building cloud infrastructures. Miles is Director of Solutions for Google Cloud; focused on delivering next-generation solutions to challenges in big data and analytics, application migration, infrastructure automation, and cost optimization. He worked as a core part of the Obama for America 2012 "TECH" team, crashed Twitter a few times, helped NASA stream the Curiosity Mars Rover landing, put Skype back online in a pinch, and plays a mean electric sousaphone.

Based in London, UK, Grace Mollison is a Cloud Solutions Architect where she helps customers to understand how to apply policies to their Google cloud platform environments as well as how to architect and deploy applications on the Google Cloud platform. In her spare time she spends time attempting to teach her international team how to speak the Queens english! Before Google Grace was a Solutions Architect at AWS where she worked with the AWS ecosystem and customers to ensure well architected solutions.

• We have awesome new intro and outro music. Did you notice?
• The thing is … Cloud IoT Core is now generally available blog site
• JupyterLab is Ready for Users blog github
• Announcing Google Cloud Spanner as a Vault storage backend blog
• How to handle mutating JSON schemas in a streaming pipeline, with Square Enix blog
• FAT* livestream

• Google Cloud Platform Solutions site
• Tutorials and Solutions site
• Machine Learning with Financial Time Series Data solution
• Implementing GCP Policies for Customer Use Cases solution
• #87 Customer Engineers with Jonathan Cham podcast
• Google Cloud Next
• Solution Architects are hiring! careers

How do I get a Docker image into Minikube without uploading it to an external registry and then downloading it all over again? Is there an easy way to do this locally?

• Minikube github
• $ docker save | (eval $(minikube docker-env) && docker load)
• Original references github Stack Overflow

Mark will be at the Game Developer's Conference | GDC in March.

We have the pleasure this week of having the Director of Solutions for Google Cloud Miles Ward and Cloud Solutions Architect Grace Mollison

In this episode, Google Play Marketing is the customer of Google Cloud Platform. Melanie and Mark chat with Dom Elliott (Google Play) and Stewart Bryson (Red Pill Analytics) about how they use our big data processing and visualisation tools to introspect what is happening in the Google play ecosystem.

Dom Elliott leads global developer marketing communications for Google Play. His goal is to help Android app and game developers improve their app quality and business performance on Google Play, by raising awareness and understanding of features that can help them find success.

Stewart Bryson is the Owner & Co-founder of Red Pill Analytics, a products and services company specializing in Cloud Analytics delivery. Red Pill is 4 years old and has about 30 employees in the US, UK and Brazil. We work with customers to accelerate their use of the public cloud for analytics, including migrating current on-premises workloads. Red Pill Analytics was engaged by Google Play to build the digital channel ingestion processes, as well as build all the Data Studio content for analyzing those channels.

• Easy distributed training with TensorFlow using tf.estimator.train_and_evaluate on Cloud ML Engine blog
• CI/CD with Less Fluff & More Awesome blog
• 96 vCPU Compute Engine instances are now generally available announcement site

• Google Play site
• Google Data Studio site docs
• Adding charts to Data Studio docs
• Google BigQuery Data Transfer Service site docs
• Google App Engine site docs
• Cloud Cloud PubSub site docs
• Cloud Functions site docs
• Google Cloud Pub/Sub Triggers docs tutorial
• Cloud Natural Language site docs

Google Play

If you want to be able to unit test your integrations with Kubernetes with client-go, how can you mock what happens inside the cluster in your unit tests?

• fake.Clientset godoc code example
• testing.Fake godoc

Melanie will be at Fat* in New York very shortly!

Mark will be at the Game Developer's Conference | GDC in March.