Labeling training data is increasingly the largest bottleneck in deploying machine learning systems. We present Snorkel, a first-of-its-kind system that enables users to train state-of-the-art models without hand labeling any training data. Instead, users write labeling functions that express arbitrary heuristics, which can have unknown accuracies and correlations. Snorkel denoises their outputs without access to ground truth by incorporating the first end-to-end implementation of our recently proposed machine learning paradigm, data programming. We present a flexible interface layer for writing labeling functions based on our experience over the past year collaborating with companies, agencies, and research laboratories. In a user study, subject matter experts build models 2.8× faster and increase predictive performance an average 45.5% versus seven hours of hand labeling. We study the modeling trade-offs in this new setting and propose an optimizer for automating trade-off decisions that gives up to 1.8× speedup per pipeline execution. In two collaborations, with the US Department of Veterans Affairs and the US Food and Drug Administration, and on four open-source text and image data sets representative of other deployments, Snorkel provides 132% average improvements to predictive performance over prior heuristic approaches and comes within an average 3.60% of the predictive performance of large hand-curated training sets.

Labeling training data is increasingly the largest bottleneck in deploying machine learning systems. We present Snorkel, a first-of-its-kind system that enables users to train state-of- the-art models without hand labeling any training data. Instead, users write labeling functions that express arbitrary heuristics, which can have unknown accuracies and correlations. Snorkel denoises their outputs without access to ground truth by incorporating the first end-to-end implementation of our recently proposed machine learning paradigm, data programming. We present a flexible interface layer for writing labeling functions based on our experience over the past year collaborating with companies, agencies, and research labs. In a user study, subject matter experts build models 2.8× faster and increase predictive performance an average 45.5% versus seven hours of hand labeling. We study the modeling tradeoffs in this new setting and propose an optimizer for automating tradeoff decisions that gives up to 1.8× speedup per pipeline execution. In two collaborations, with the U.S. Department of Veterans Affairs and the U.S. Food and Drug Administration, and on four open-source text and image data sets representative of other deployments, Snorkel provides 132% average improvements to predictive performance over prior heuristic approaches and comes within an average 3.60% of the predictive performance of large hand-curated training sets.

No abstract

Populating large-scale structured databases from unstructured sources is a critical and challenging task in data analytics. As automated feature engineering methods grow increasingly prevalent, constructing sufficiently large labeled training sets has become the primary hurdle in building machine learning information extraction systems. In light of this, we have taken a new approach called data programming [7]. Rather than hand-labeling data, in the data programming paradigm, users generate large amounts of noisy training labels by programmatically encoding domain heuristics as simple rules. Using this approach over more traditional distant supervision methods and fully supervised approaches using labeled data, we have been able to construct knowledge base systems more rapidly and with higher quality. Since the ability to quickly prototype, evaluate, and debug these rules is a key component of this paradigm, we introduce DDLite, an interactive development framework for data programming. This paper reports feedback collected from DDLite users across a diverse set of entity extraction tasks. We share observations from several DDLite hackathons in which 10 biomedical researchers prototyped information extraction pipelines for chemicals, diseases, and anatomical named entities. Initial results were promising, with the disease tagging team obtaining an F1 score within 10 points of the state-of-the-art in only a single day-long hackathon's work. Our key insights concern the challenges of writing diverse rule sets for generating labels, and exploring training data. These findings motivate several areas of active data programming research.

No abstract