Snorkel AI is a Gartner Cool Vendor for Data-Centric AI.

Get the report
Image
  • Product
    • PRODUCT
    • Menu Item
      • Snorkel Flow – Accelerate AI development with the data-centric platform powered by programmatic labeling.
      • Data labeling – Label programmatically by distilling expertise into functions that power intelligent auto-labeling.
      • Model training and analysis – Continuously update and analyze models to guide rapid iteration and improvement.
    • Menu Item
      • Foundation models – Bridge adaptation and deployment gaps to use foundation models for enterprise production use cases.
      • Enterprise – Benefit from enterprise-grade interoperability, security, expertise, and more.
  • Solutions
    • Solutions
    • Menu Item
      • Selected industry use cases
      • AI for banking – Personalize customer interactions, manage risk, and improve resource utilization.
      • AI for healthcare – Speed clinical trial success, improve patient outcomes, and enhance research.
      • AI for government – Build machine learning models and AI applications across a wide variety of missions and use cases.
      • AI for insurance – Detect fraud, speed claims processing, and improve underwriting workflows.
      • AI for telecom – Assess network health, tailor customer support, and detect security risks.
    • Menu Item
      • Selected AI applications
      • Document classification – Improve performance by exploiting features unique to your data with custom classification apps.
      • Named entity recognition – Solve domain-specific syntactic and semantic challenges with precise NER apps.
      • Information extraction – Collect useful text and data from virtually any table or form with flexible extraction apps.
      • Conversational AI – Automate responses to customer inquiries to improve efficiency and customer satisfaction.
      • Sentiment analysis – Tackle complex NLP challenges with nuanced sentiment analysis apps.
  • Technology
    • TECHNOLOGY
    • Menu Item
      • Data-centric AI – Learn about the impact of moving from model-centric to data-centric AI.
      • Weak supervision – Explore weak supervision approaches and how they accelerate training data creation.
    • Menu Item
      • Programmatic labeling – See how programmatic labeling breaks through the primary bottleneck facing AI.
      • Snorkel research project – Read how Snorkel open source has advanced to production as the Snorkel Flow platform.
  • Case studies
    • CASE STUDIES
    • Menu Item
      • Apple
      • Big four consulting firm
      • Fortune 50 bank
      • Fortune 500 telecom
      • Global custodian bank
      • Global financial services leader
      • Global insurance leader
      • Google
    • Menu Item
      • Intel
      • Memorial Sloan Kettering Cancer Center
      • Schlumberger
      • Stanford Medicine
      • Tide
      • Top 5 Pharma
      • Top US bank
  • Resources
    • RESOURCES
    • Menu Item
      • Blog – Browse the latest posts from Snorkel AI experts, customers, and industry leaders.
      • Events – Attend events to discover new ideas and make connections with AI/ML practitioners.
      • Frequently asked questions – Explore frequently asked questions on scalable AI development, Snorkel AI, and Snorkel Flow
    • Menu Item
      • Press – Read the latest news about Snorkel AI and Snorkel Flow customers.
      • Research papers – Explore 60+ peer-reviewed papers published as part of the Snorkel research project.
  • Company
    • COMPANY
    • Menu Item
      • About us – Snorkel AI is redefining how AI applications are built. Learn about our mission, team, and culture.
      • Careers – We’re bringing together some of the best minds in AI and machine learning. Join us.
      • Partners – We work with an innovative ecosystem of partners focused on delivering value through data-centric AI.
  • Request a demo
Request a demo

Semi-Supervised Aggregation of Dependent Weak Supervision Sources with Performance Guarantees

Roberto Iriondo July 18, 2021

This work shows a rigorous technique for efficiently selecting small subsets of the labelers so that a majority vote from such subsets has a provably low error rate.

Fast and Three-Rious: Speed up Weak Supervision With Triplet Methods

11450pwpadmin November 20, 2020

Introducing FlyingSquid, a weak supervision framework that runs orders of magnitude faster than previous weak supervision approaches and requires fewer assumptions

Train and You’ll Miss It: Interactive Model Iteration With Weak Supervision…

11450pwpadmin November 13, 2020

This paper provides a series of results studying how performance scales with changes in source coverage, source accuracy, and the Lipschitzness of label distributions in the embedding space, and compare this rate to standard weak supervision.

The Role of Massively Multi-Task and Weak Supervision in Software 2.0

11450pwpadmin December 17, 2019

Outlining a vision for a Software 2.0 lifecycle centered around the idea that labeling training data can be the primary interface to Software 2.0 systems.

Snuba: Automating Weak Supervision to Label Training Data

11450pwpadmin December 16, 2019

Presenting Snuba, a system to automatically generate heuristics using a small labeled dataset to assign training labels to a large, unlabeled dataset in the weak supervision setting.

Osprey: Weak Supervision of Imbalanced Extraction Problems Without Code

11450pwpadmin December 12, 2019

Proposing Osprey, a weak-supervision system suited for highly imbalanced data, built on top of the Snorkel framework.

Interactive Programmatic Labeling for Weak Supervision

11450pwpadmin December 8, 2019

Demonstrating in synthetic and real-world experiments how two simple labeling function acquisition strategies outperform a random baseline.

Bootstrapping Conversational Agents with Weak Supervision

11450pwpadmin December 7, 2019

This paper presents a framework called search, label, and propagate (SLP) for bootstrapping intents from existing chat logs using weak supervision.

Software 2.0 and Snorkel: Beyond Hand-Labeled Data

11450pwpadmin December 19, 2018

This paper describes Snorkel, a system that enables users to help shape, create, and manage training data for Software 2.0 stacks.

Socratic Learning: Augmenting Generative Models to Incorporate Latent Subsets in Training Data

11450pwpadmin November 13, 2017

Introducing Socratic learning, a paradigm that uses feedback from a discriminative model to automatically identify latent data subsets in training data.

Image

Product

  • Snorkel Flow
  • Label
  • Model
  • Foundation models
  • Enterprise

Solutions

AI applications

  • Document classification
  • Named entity recognition
  • Information extraction
  • Conversational AI
  • Sentiment analysis

Industry use cases

  • AI for banking
  • AI for government
  • AI for healthcare
  • AI for insurance
  • AI for telecom

Technology

  • Snorkel AI Technology
  • Data-centric AI
  • Weak supervision
  • Programmatic labeling
  • Snorkel research project

Case studies

  • Apple
  • Big four consulting firm
  • Fortune 50 bank
  • Fortune 500 telecom
  • Global custodian bank
  • Global financial services leader
  • Global insurance leader
  • Google
  • Intel
  • Memorial Sloan Kettering Cancer Center
  • Stanford Medicine
  • Tide
  • Top 5 biotech
  • Top US bank

Resources

  • Blog
  • Events
  • FAQ
  • Press
  • Research papers
  •  

 

Company

  • Overview
  • Careers
  • Snorkel logo

Contact

  • Contact us
  • Request a demo

Copyright © 2022 Snorkel AI, Inc. All rights reserved.
Terms of Use Privacy Cookie Policy
Image
Image
Image