Model-to-World Lab · Stanford MS&E
M2W studies how decisions built in mathematical, simulated, and data-derived worlds perform when they meet reality. We build the world, compute the decision, and measure the gap.
The M2W research loop
The model-to-world gap is not a final disclaimer. It is examined at each stage and updated with evidence from reality.
Represent the mechanisms, constraints, and uncertainty that matter for a consequential decision.
Explore ordinary, rare, shifted, and counterfactual worlds before acting in the real one.
Compare policies and optimize performance while accounting for uncertainty and misspecification.
Test what survives the model-to-world gap, then validate and recalibrate with new evidence.
Examples · M2W Decision Labs
Interactive examples make the loop inspectable: model the mechanism, learn from partial evidence, protect the decision, and name what must be validated next.
Blanchet Research Group
M2W is led by José H. Blanchet, Professor of Management Science & Engineering at Stanford. It brings together two decades of research in stochastic simulation, model risk, optimization, learning, and causal inference around one question: how do we make decisions that survive contact with the world?