By Mazin Gilbert, AT&T
Artificial Intelligence (AI) has come a long way since its debut in Bell Labs in the 1950s. It’s transformed from a sci-fi concept to now being a reality in our data-driven world. And while the technology and applications may have changed over the decades, the underlying goal remains the same – enhance and simplify our everyday lives.
AI is expected to impact product development and innovation across all industries. Needless to say, these advancements wouldn’t be possible without people constantly working to solve problems, think outside the box, innovate and create forward-looking solutions.
But to advance AI faster, we needed to break down barriers and make it more accessible to developers, entrepreneurs and businesses.
We unveiled the Acumos AI Project nearly a year ago. We collaborated with Tech Mahindra to build Acumos, which is an open source AI platform under The Linux Foundation Deep Learning – an umbrella including 15 corporate members. And, we recently co-sponsored the first-ever Acumos AI Challenge with Tech Mahindra. We challenged students, developers and data scientists across the country to come up with AI solutions. The contest featured a total of $100,000 in cash awards.
We called for innovative, fresh ideas – and the industry answered. More than 300 developers participated, and submissions spanned industries including healthcare, media and entertainment, and security.
Entries were judged against 5 judging criteria categories:
- Novelty & Originality
- Viability & Impact on Market
- Difficulty of Technical Implementation
- Best Supports Package Requirements & Intended Function
We chose 3 finalists and invited them to present their ideas on stage at AT&T Spark in San Francisco last week. These are the innovators leading the way in AI development. See below for what they came up with.
Acumos AI Challenge Winner: Chris Buonocore
Chris took home $50,000 with his Acumos Property Assistant. This is a machine learning model for pricing your home based on fundamental property characteristics. Acumos Property Assistant presents an unbiased, transparent way to accurately price property and help simplify the complex process of selling a home. Congratulations, Chris!
Meet Chris and learn more about the Acumos property assistant in this video.
Acumos AI Challenge Finalist: Rhutvij Savant
Rhutvij claimed $25,000 for his Cancer Classification solution. The model, which classifies tumors as malignant or benign breast cancer, could provide more accurate prognosis and ultimately better care – potentially saving lives.
Acumos AI Challenge Finalists: Jinhe Shi and Yuhua Gong
Jinhe Shi and Yuhua Gong were awarded $25,000 for their User Review Prediction model. This solution is used to classify user reviews into positive and negative, giving business owners valuable insights to help improve service.
Not only did these finalists walk out with prize money and serious bragging rights, they also gained exposure and experience during AT&T Spark. AT&T’s invite-only innovation showcase offered the inventors a captive audience filled with industry leaders, fellow developers, and a host of reporters and analysts. Check out some highlights from the day for yourself in this video.
We want to thank all who entered the Acumos AI Challenge. We were impressed and inspired by the creative solutions we saw.
The challenge exemplified how open collaboration will help drive innovative solutions faster and more efficiently.
If you have an idea to bring to life, submit it to the Acumos AI project. Whether you’re a data scientist, developer, teacher or student – you can participate and influence the AI revolution.
This was originally posted at https://about.att.com/story/2018/acumos__challenge_winner.html