# Learning under random distributional shifts

arXiv: https://arxiv.org/abs/2306.02948 · HTML: https://mlsanity.com/paper/2306.02948/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 60 references that are in the map.
- First posted on arXiv 5 Jun 2023.

## Builds on (most cited first, 20 of 20)

- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [Unsupervised Domain Adaptation by Backpropagation](https://mlsanity.com/paper/1409.7495/) (2014)
- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [In Search of Lost Domain Generalization](https://mlsanity.com/paper/2007.01434/) (2020)
- [Unlabeled Data Improves Adversarial Robustness](https://mlsanity.com/paper/1905.13736/) (2019)
- [Wild Patterns: Ten Years After the Rise of Adversarial Machine Learning](https://mlsanity.com/paper/1712.03141/) (2017)
- [Using Pre-Training Can Improve Model Robustness and Uncertainty](https://mlsanity.com/paper/1901.09960/) (2019)
- [Learning Models with Uniform Performance via Distributionally Robust Optimization](https://mlsanity.com/paper/1810.08750/) (2018)
- [Understanding and Mitigating the Tradeoff Between Robustness and Accuracy](https://mlsanity.com/paper/2002.10716/) (2020)
- [Understanding Self-Training for Gradual Domain Adaptation](https://mlsanity.com/paper/2002.11361/) (2020)
- [Anchor regression: heterogeneous data meets causality](https://mlsanity.com/paper/1801.06229/) (2018)
- [Self-training Avoids Using Spurious Features Under Domain Shift](https://mlsanity.com/paper/2006.10032/) (2020)
- [Domain adaptation under structural causal models](https://mlsanity.com/paper/2010.15764/) (2020)
- [Targeting for long-term outcomes](https://mlsanity.com/paper/2010.15835/) (2020)
- [Fundamental Tradeoffs in Distributionally Adversarial Training](https://mlsanity.com/paper/2101.06309/) (2021)
- [Dynamic Placement in Refugee Resettlement](https://mlsanity.com/paper/2105.14388/) (2021)
- [Outcome-Driven Dynamic Refugee Assignment with Allocation Balancing](https://mlsanity.com/paper/2007.03069/) (2020)
- [Calibrated inference: statistical inference that accounts for both sampling uncertainty and distributional uncertainty](https://mlsanity.com/paper/2202.11886/) (2022)
- [Distributionally robust and generalizable inference](https://mlsanity.com/paper/2209.09352/) (2022)

## Built on by (most cited first, 6 of 6)

- [Beyond Reweighting: On the Predictive Role of Covariate Shift in Effect Generalization](https://mlsanity.com/paper/2412.08869/) (2024)
- [Optimal Empirical Risk Minimization under Temporal Distribution Shifts](https://mlsanity.com/paper/2507.13287/) (2025)
- [Diagnostics for Individual-Level Prediction Instability in Machine Learning for Healthcare](https://mlsanity.com/paper/2603.00192/) (2026)
- [Robustness of Refugee-Matching Gains to Off-Policy Evaluation Choices](https://mlsanity.com/paper/2605.06686/) (2026)
- [Harnessing non-adversarial robustness in large language models](https://mlsanity.com/paper/2605.29816/) (2026)
- [Augmented Inverse Hybrid Weighting: Robust Inference under Deterministic and Random Distribution Shifts](https://mlsanity.com/paper/2608.00701/) (2026)

## Abstract (arXiv)

Many existing approaches for generating predictions in settings with distribution shift model distribution shifts as adversarial or low-rank in suitable representations. In various real-world settings, however, we might expect shifts to arise through the superposition of many small and random changes in the population and environment. Thus, we consider a class of random distribution shift models that capture arbitrary changes in the underlying covariate space, and dense, random shocks to the relationship between the covariates and the outcomes. In this setting, we characterize the benefits and drawbacks of several alternative prediction strategies: the standard approach that directly predicts the long-term outcome of interest, the proxy approach that directly predicts a shorter-term proxy outcome, and a hybrid approach that utilizes both the long-term policy outcome and (shorter-term) proxy outcome(s). We show that the hybrid approach is robust to the strength of the distribution shift and the proxy relationship. We apply this method to datasets in two high-impact domains: asylum-seeker assignment and early childhood education. In both settings, we find that the proposed approach results in substantially lower mean-squared error than current approaches.