Sam WeissAutomating Dataset Labeling with LLMs for Supervised Topic ModelsIn this post, I explore how to use large language models (LLMs) to automate dataset labeling for building supervised topic models.Sep 4Sep 4
Sam WeissEnhancing Diversity in RAG Document Retrieval Using Projection-Based TechniquesIntroductionAug 211Aug 211
Sam WeissinBuilding IbottaHow to Use Previous Model Deployments to Build (a Better) Causal ModelDo you experiment with different models in production? Read on to see how you can use observational causal inference to build a better…Jun 23, 2021Jun 23, 2021
Sam WeissinTowards Data ScienceMR-Uplift: Multiple Responses in Uplift ModelsA package to build Uplift (or heterogeneous treatment effects) models that builds and evaluates tradeoffs with multiple responsesApr 30, 20202Apr 30, 20202
Sam WeissinBuilding IbottaThe MR-Uplift PackageEvaluating Multiple Responses in Uplift ModelsApr 23, 2020Apr 23, 2020
Sam WeissinBuilding IbottaEstimating and Visualizing Business Tradeoffs in Uplift ModelsUplift modeling is a powerful tool to assign treatments to optimize for a particular response variable. However, we often face tradeoffs…Nov 15, 2019Nov 15, 2019
Sam WeissinBuilding IbottaMaximizing The ERUPT Metric for Uplift ModelsIntroduction:Aug 6, 2019Aug 6, 2019
Sam WeissinBuilding IbottaERUPT: Expected Response Under Proposed TreatmentsIntroduction:May 13, 20191May 13, 20191
Sam WeissinBuilding IbottaMultiple Responses in A/B TestingHow Ibotta data scientists build and use priors on the relationships of treatment effects to get more accurate A/B test results.Oct 16, 2018Oct 16, 2018
