Practical Marketplace Optimization at Uber Using Causally-Informed Machine Learning
This work addresses a critical business challenge for Uber in efficiently managing budgets across cities, though it appears incremental as it builds on existing causal ML methods.
The paper tackles the problem of optimizing budget allocation for marketplace levers like driver incentives and rider promotions at Uber by developing an end-to-end machine learning and optimization procedure, resulting in substantial improvements in resource allocation efficiency.
Budget allocation of marketplace levers, such as incentives for drivers and promotions for riders, has long been a technical and business challenge at Uber; understanding lever budget changes' impact and estimating cost efficiency to achieve predefined budgets is crucial, with the goal of optimal allocations that maximize business value; we introduce an end-to-end machine learning and optimization procedure to automate budget decision-making for cities, relying on feature store, model training and serving, optimizers, and backtesting; proposing state-of-the-art deep learning (DL) estimator based on S-Learner and a novel tensor B-Spline regression model, we solve high-dimensional optimization with ADMM and primal-dual interior point convex optimization, substantially improving Uber's resource allocation efficiency.