Come hear about TensorFlow models built and trained by your colleagues in the Boston developer community. We have 4 very interesting talks summarized below with speaker bios. The talks will focus on practice rather than theory with blow-by-blow accounts from hands-on practitioners
SMALL MOLECULE DRUG DISCOVERY:
Reverie Labs approaches small molecule design with machine learning, backed by Tensorflow, at its core. We are harnessing the pattern-recognition capacity of deep neural networks to build rich computational models for molecules. In this talk, we will share how deep neural networks can be used to represent and model chemical structures, and explain how models developed for images, video, and text can be applied to drug discovery.
Ankit Gupta is a co-founder of Reverie Labs, which uses machine learning to accelerate drug discovery. Ankit graduated from Harvard with a B.A. in Computer Science and an M.S. in Computer Science.
CREATING AN NCAA BASKETBALL MARCH MADNESS:
Ever wondered how we can use Machine Learning to create an NCAA Basketball March Madness Bracket? What variables will you even be looking at? Where will you get past and current data about players, teams, venues, etc? Lastly, with all the data how will you devise an algorithm to make a prediction? Join me in walking through code on how I was able to achieve this.
Praveen Aravamudham is currently the Chief Technology Officer at Berkshire Hathaway Shoe Holdings Group. He has over 15 years software engineering industry experience in designing and implementing high-performance distributed software systems in various programming languages. Praveen is a hackathon “geek” ready to join a team, and frequently a contributor to a winning team. He mentors young developers and enjoys conversation about technology past/present/future.
AIY VISION KIT – EMBEDDED ML FOR STEM AND MAKERS
Google’s “AI Yourself” Vision Kit available at Target Stores is a low-cost, developer friendly development kit aimed at makers and STEM students. It includes pre-trained models for detecting mood, pets, meals, with the ability to load your own custom model.
Chad Hart is a frequent speaker, maker, blogger at webrtcHacks.com, and event organizer with Kranky Geek and WebRTC Boston. He is a product management, marketing and strategy consultant with cwh.consulting. , Chad recently ran an enterprise new product incubator program with the lates latest speech analytics, voice assistant, and telephony technologies.
Medical imaging introduces unique challenges to the development of machine learning algorithms. From volumetric data to small datasets to potentially high inter-rater variability, training an accurate model often requires overcoming a series of obstacles not typically faced by those working with natural images. In this talk, we'll discuss some common difficulties and how careful planning combined with strategic network design can result in accurate, impactful models.
Neil Tenenholtz leads the machine learning and software engineering efforts at the MGH & BWH Center for Clinical Data Science, where his responsibilities include the training of novel deep learning models for clinical diagnosis, the development of robust infrastructure for their deployment into the clinical setting, and the creation of tooling to facilitate these processes. Prior to joining the center, Neil was a Senior Research Scientist at Fitbit where he leveraged machine learning and modeling techniques to develop new features and algorithms that reside both on-device and in the cloud.
Registration does not guarantee your seat. Running high-quality free events like this is challenging because we have to guess the percentage of registrations will show.