mlsanityData to 7 Oct 2026

The Value of Prediction in Identifying the Worst-Off

arXiv 2501.19334 ↗First posted 31 Jan 2025cs.CY cs.LG stat.MLcited by 5

Not on the map. Why?

Abstract (arXiv)

Machine learning is increasingly used in government programs to identify and support the most vulnerable individuals, prioritizing assistance for those at greatest risk over optimizing aggregate outcomes. This paper examines the welfare impacts of prediction in equity-driven contexts, and how they compare to other policy levers, such as expanding bureaucratic capacity. Through mathematical models and a real-world case study on long-term unemployment amongst German residents, we develop a comprehensive understanding of the relative effectiveness of prediction in surfacing the worst-off. Our findings provide clear analytical frameworks and practical, data-driven tools that empower policymakers to make principled decisions when designing these systems.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 11 of its 36 references that are in the map. First posted on arXiv 31 Jan 2025.

Builds on

11 of 36 references.

References · 11

  1. 2017Smart "Predict, then Optimize"cited by 332
  2. 2023Difficult Lessons on Social Prediction from Wisconsin Public Schoolscited by 35
  3. 2022A Validity Perspective on Evaluating the Justified Use of Data-driven Decision-making Algorithmscited by 30
  4. 2021Causal Decision Making and Causal Effect Estimation Are Not the Same... and Why It Matterscited by 25
  5. 2019More Efficient Policy Learning via Optimal Retargetingcited by 23
  6. 2023The Relative Value of Prediction in Algorithmic Decision Makingcited by 14
  7. 2023Counterfactual Prediction Under Outcome Measurement Errorcited by 10
  8. 2024Allocation Requires Prediction Only if Inequality Is Lowcited by 8
  9. 2024Learning treatment effects while treating those in needcited by 8
  10. 2024Scarce Resource Allocations That Rely On Machine Learning Should Be Randomizedcited by 7
  11. 2024Evaluating the Effectiveness of Index-Based Treatment Allocationcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026MLReplicate: Benchmarking Autonomous Research Systems for Machine Learning Reproducibilitycited by 6
  2. 2026Good Allocations from Bad Estimatescited by 2
  3. 2026Performative Learning Theorycited by 0
  4. 2026On the Meta-Design of Allocation Problemscited by 0
  5. 2026The Limits of AI-Driven Allocation: Optimal Screening under Aleatoric Uncertaintycited by 0