Method
Weak supervision
Programmatic or heuristic labels combined into training data.
9 cases from 9 companies apply Weak supervision. 2 were checked in an online A/B test; 2 report results in production.
Typical tasks: Catch fraud and abuse, Detect anomalies, Rank search results, Moderate content at scale and 4 more. Companies: Stripe, Zillow, Samokat.tech, Celsus, DoorDash, Netflix and 3 more. Often combined with: Threshold selection, Multi-label classification, Autoencoder anomaly detection, Isolation Forest, Gradient boosting and 35 more.
A Machine Learning Model for Detecting Atypical Bank Customer Transactions![]() | ![]() | banking | 2025-05 |
| The ML flywheel: How we continually improve our models to reduce card testing | fintech | 2024-12 | |
Category Prediction for Search Query Understanding![]() | ![]() | e-commerce | 2024-04 |
| Navigating Fair Housing Guardrails in LLMs | marketplace | 2024-01 | |
NER for Marketplace Search Queries![]() | ![]() | marketplace | 2023-05 |
Modeling the Behavior of an Oil Spill![]() | ![]() | manufacturing | 2023-05 |
Data-Centric AI: A Review of Methods![]() | ![]() | healthtech | 2023-05 |
| How DoorDash Upgraded a Heuristic with ML to Save Thousands of Canceled Orders | food-delivery | 2023-01 | |
| Machine Learning for Fraud Detection in Streaming Services | media-streaming | 2022-09 |




