Presented By O’Reilly and Intel Nervana
Put AI to work
September 17-18, 2017: Training
September 18-20, 2017: Tutorials & Conference
San Francisco, CA

Deep learning with limited labeled data

Avesh Singh (Cardiogram), Brandon Ballinger (Cardiogram)
1:45pm–2:25pm Tuesday, September 19, 2017
Implementing AI
Location: Yosemite BC Level: Intermediate
Secondary topics:  Data and training, Medicine
Average rating: ****.
(4.50, 2 ratings)

Prerequisite Knowledge

  • A basic understanding of machine learning and neural networks

What you'll learn

  • Learn to train deep neural networks with limited amounts of labeled data
  • Discover practical debugging techniques to figure out why your neural network isn't learning

Description

Deep learning is fueled by large labeled datasets, but in domains like medicine, each label represents a human life at risk. Avesh Singh and Brandon Ballinger offer an overview of autoencoders, heuristic training, and few-shot learning, with an emphasis on practical tips to create high-performing models utilizing hundreds of thousands of unlabeled data points and only thousands of labeled points.

Unsupervised, semisupervised, and one-shot learning techniques let us supplement scarce labels with vast quantities of unlabeled data. Avesh and Brandon walk you through a large-scale (N=10,000) study with UCSF Cardiology that predicts atrial fibrillation, the most common heart arrhythmia and the number-one predictor of stroke, using heart rate measurements from an Apple Watch, utilizing labeled and unlabeled data to identify episodes with an AUC of 0.88.

Photo of Avesh Singh

Avesh Singh

Cardiogram

Avesh Singh is an engineer at Cardiogram, a startup that applies deep learning to wearable data. Previously, Avesh worked at Nest Labs and Google. He holds a a BS and MS in computer science from Carnegie Mellon University.

Photo of Brandon Ballinger

Brandon Ballinger

Cardiogram

Brandon Ballinger is a cofounder at Cardiogram. Previously, Brandon was a cofounder at Sift Science and an engineer at Google on speech recognition and ads quality. He was also one of the engineers called in by the White House to help fix Healthcare.gov. Brandon holds a BS in computer science from the University of Washington.

Leave a Comment or Question

Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?

Join the conversation here (requires login)

Comments

Brandon Ballinger | CO-FOUNDER
09/09/2017 12:46pm PDT

Hi Xu! For recommended background—we’ll assume knowledge of deep learning in general (including convolutional neural networks and LSTMs). It’ll be helpful if you’ve played with TensorFlow, Keras, etc., since we’ll be showing code samples.

We won’t assume any background in the semi-supervised techniques, few-shot learning, etc.—that’s what we’re hoping you walk away understanding after this talk!

Xu Zhang | MEMBER OF RESEARCH STAFF
08/24/2017 9:42am PDT

Any recommended papers/articles to read before attending this talk?