How We Share the Latest AI & ML Developments Within Microsoft
We recently concluded the Fall 2018 edition of the Machine Learning, AI & Data Science (MLADS) conference, Microsoft’s largest internal gathering of employees focused specifically on these areas. This latest edition was the eighth in a popular series that we launched back in 2014. Over 3,500 employees tuned into the sold-out conference, both in person in Redmond and over livestream throughout the world, and thousands more will tune into MLADS session recordings over coming weeks and months.
As application of AI and ML explode both within Microsoft and in our external products and services, the growth of our community interest groups catering to these areas has been very rapid. The MLADS conference itself is unique in that it is almost entirely driven by enthusiastic community volunteers – a band of employees unified in its passion for AI and ML, and a desire to network and learn from one another. The “call for content” that goes out for this conference series routinely gets several hundreds of submissions, and our volunteer team helps triage these submissions and curate the best ones for our event.
The fall 2018 conference featured over 95 talk sessions, 20 tutorials and 65 poster/demo sessions covering a gamut of topics including Azure Machine Learning, advanced Deep Learning, Generative Adversarial Networks, Natural Language Processing, Text Mining, Vision, Speech, Cognitive Services, Time Series Methods, Forecasting, Azure Batch AI, Design & Interpretability, the very latest AI tools, and numerous applications of AI – now even in the generation of original music and creation of original art.
Poster/Demo Reception at the Fall 2018 Microsoft MLADS conference
Talks and tutorials about open source offerings from Microsoft and third parties are an integral part of the conference, and this time was no exception.
A panel of volunteer judges help us shortlist top submissions for our Distinguished Contribution Awards, an internal recognition program for top-class work being done in this field. Selective submissions are also chosen for inclusion in the prestigious [internal] Microsoft Journal of Applied Research.
The conference opened to packed keynote sessions by Joseph Sirosh, Corporate Vice President of Cloud AI, and Todd Holmdahl, Corporate Vice President of Quantum Computing. Joseph’s keynote, on the AI Tidal Wave, highlighted the impact AI can have (and is having already) on the lives of people all around the world. His keynote featured leading-edge demos showing some of the latest AI and ML-led innovations happening across Microsoft.
Joseph Sirosh, Corporate Vice President of Cloud AI, Microsoft
As part of Todd’s keynote, about Empowering the Quantum Revolution, we pre-announced the Microsoft Quantum Development Kit (QDK), a set of development tools for quantum developers. These tools were subsequently made available to the public, visit this web page for more information.
Todd Holmdahl, Corporate Vice President of Quantum Computing, Microsoft
The Fall 2018 MLADS conference had several other notable features, including:
- A set of introductory talks and tutorials, to get newcomers caught up on the subject. This set of sessions and a set of associated programs helps developers who are not yet formally trained in ML and AI quickly ramp up into this space and get their hands dirty by doing real projects.
- A set of immensely popular sessions from colleagues who joined the Microsoft community relatively recently, including from LinkedIn (based in Silicon Valley) and Maluuba (based in Montreal).
- A Startup Showcase featuring the latest set of exciting startups from the Microsoft Accelerator program in Seattle.
- A Robotics Expo, featuring many new ideas and experiments involving the latest in ML and robotics.
- Informal lunchtime meetups on a variety of topics, including ML for sports predictions, ethics and AI, AI in neuroscience, and AI for Earth.
- A closing plenary session on Intelligent Design / Design for Intelligence, by Bill Buxton.
The fall event also featured our first Family Build-a-Robot Night – children from many Microsoft families came together to build and program wheeled robots, which they then then tested on a variety of tracks. This was just as much fun for our teachers and volunteers as the kids themselves!
Between bi-annual MLADS conferences, we also have a monthly set of community events and periodic hackathons.
These investments are helping our fast-growing (and global) group of AI and ML practitioners learn from each another, even as we continue to enhance Microsoft’s products through the power of everyday AI.
ML Blog Team
Follow us on Twitter