By Charis Loveland, Senior Program Manager for Cortana Intelligence Competitions in the Data Group at Microsoft.
Machine Learning algorithms powered by intelligent applications serve useful functions in our daily lives in ways we may not even be aware of. For instance, predictive analytics allow businesses to retain key customers, help assembly lines and buildings to run more efficiently, and help us find movies that we are likely to find intriguing. The ML field has gained tremendous traction and respect over the last decade, prompting Harvard Business Review to name the Data Scientist the sexiest job of the 21st century.
To encourage new ML applications and foster a vibrant online community, we are thrilled to launch Cortana Intelligence Competitions, a gamification feature of Cortana Intelligence Suite, as well as our first competition Decoding Brain Signals. This platform provides an intuitive and fun environment to hone users’ data science and analytics expertise, and our first competition will allow you to have the chance to contribute to the important field of neuroscience to win prizes and recognition.
Competitions allow you to:
- Explore unique data sets by participating in varying levels of competitions. These datasets are being released publically for the first time, so don’t miss out.
- Compete with data science experts at a global level. What better way to learn than some healthy competition? Can you build predictive models with a higher accuracy than the experts? We allow you to submit multiple solutions, so keep learning and trying.
- Advance the data science field, and in many cases science in general, by contributing your ideas and creativity toward meaningful challenges and sharing results with other experts.
- Win prizes worth thousands of dollars or find yourself on the coveted Top 10 public leaderboard. Here’s your opportunity for fame and glory!
You do not need to be an expert to compete. In fact, I joined Microsoft two months ago with no background in data science, and I’ve already participated in several internal competitions. Our tutorials, videos, and data set descriptions make it easy for those with an interest in analytics and data science to compete.
You’ll be able to submit your first competition entry in four easy steps:
- Find the competition you’d like to participate in the Cortana Intelligence Gallery. Then click on the Enter Competition button to copy the Starter Experiment into your existing Azure ML workspace. You can create a free workspace without a credit card by simply logging in with a valid Microsoft account or Office 365 account. Add your special sauce here using either built-in modules or by bringing your R/Python scripts directly, to improve on the target performance metric.
- Create a Predictive Experiment with the trained model out of your Starter Experiment, then adjust the input and output schema of the web service to ensure they conform to the specification from the Competition documentation.
- Deploy a web service out of your Predictive Experiment. Test your web service using the Test button or the Excel template automatically created for you to ensure it is working properly.
- Submit your web service as the competition entry, and see your public score in the Cortana Intelligence Gallery competition page. And celebrate if you make into the leaderboard!
After you successfully submit an entry, you can go back to the copied Starter Experiment, iterate, and update your Predictive Experiment, update the web service, and submit a new entry.
Azure ML Studio provides a rich set of ML modules as well as data processing modules in a friendly GUI for constructing experiments. It also allows experienced data scientists to bring custom R and/or Python scripts for native execution. The R and Python runtime in the Studio come with a rich set of open source R/Python packages, and additional packages can be imported as script bundles and referenced in the scripts as well.
The Competition supplies a Starter Experiment that leverages many built-in modules in Azure ML, but you can choose to replace them with R or Python scripts constructed externally and imported into the Azure ML environment to create a valid entry for submission.
Azure ML Studio also has a built-in JuPyteR Notebook service for you to do freestyle data exploration. Of course, you can always download the datasets used in the Competition and explore it locally in your favorite tool.
For a complete set of rules and features, please check out our FAQ. We’ve prepared an exciting competition launching with the release of this feature, and it is open for participation. Enter today.
Decoding Brain Signals
Prizes: $5K ($3K, $1.5K, $0.5K)
Timeline: 30th March – 30th June 2016 (3 months)
Topic: The inaugural competition is based on one of the greatest opportunities for 21st-century science – decoding signals from the human brain. Each year, millions of people worldwide suffer from brain related disorders and injuries. Neurosurgeons and computer scientists have been partnering to understand how our brain interprets electric signals to advance the treatment opportunities available for individuals with brain injuries. The medical community and ultimately, patients will benefit from your expertise in ML and data science to help decode these signals.
Through this competition, you will play a key role in bringing the next generation of care to patients through advancing neuroscience research. Build the most intelligent model and accurately predict the image shown to a person (either a house or a face) based on electric signals in the brain. The Grand Prize winner will get $3,000 cash, followed by a 2nd prize of $1,500, and a 3rd prize of $500.
We look forward to seeing you contribute to scientific advances and hone your skills with this exciting first dataset.
If you want to get some practice first to learn the end-to-end workflow of constructing and submitting a competition entry, please visit our Iris multiclass classification practice contest.
Please contact us with any questions at firstname.lastname@example.org.
We wish you the best of luck!