2026
- Unai Fischer-Abaigar, Emily Aiken, Christoph Kern, and Juan Carlos PerdomoUnder Review, 2026
There is an extensive literature that studies how to find optimal policies in resource allocation problems, taking the underlying design parameters that define the allocation, such as what data is collected, how many people can be served, and quality of service as fixed constraints. Yet, from a planner’s perspective, these design parameters are themselves optimization variables that are just as important in determining overall welfare as selecting the optimal targeting rule for a given set of constraints. This realization motivates a rich set of meta-design questions exploring how planners should make principled decisions about investments in prediction, capacity constraints, and treatment quality, all of which lie upstream of classical policy optimization. Building on initial theoretical work in this space, our paper has three main contributions. First, we formally define the broad meta- design space of resource allocation problems. Second, we develop empirical tools that enable practitioners to reliably navigate it. Third, we demonstrate the framework in two real-world case studies on German employment services and targeted cash transfer programs in Ethiopia.
- Julian Rodemann, Unai Fischer-Abaigar, James Bailie, and Krikamol MuandetIn International Conference on Machine Learning (ICML), 2026
Performative predictions influence the very outcomes they aim to forecast. We study performative predictions that affect a sample (e.g., only existing users of an app) and/or the whole population (e.g., all potential app users). This raises the question of how well models generalize under performativity. For example, how well can we draw insights about new app users based on existing users when both of them react to the app’s predictions? We address this question by embedding performative predictions into statistical learning theory. We prove generalization bounds under performative effects on the sample, on the population, and on both. A key intuition behind our proofs is that in the worst case, the population negates predictions, while the sample deceptively fulfills them. We cast such self-negating and self-fulfilling predictions as min-max and min-min risk functionals in Wasserstein space, respectively. Our analysis reveals a fundamental trade-off between performatively changing the world and learning from it: the more a model affects data, the less it can learn from it. Moreover, our analysis results in a surprising insight on how to improve generalization guarantees by retraining on performatively distorted samples. We illustrate our bounds in a case study on prediction-informed assignments of unemployed German residents to job trainings, drawing upon administrative labor market records from 1975 to 2017 in Germany.
2025
- Unai Fischer-Abaigar, Christoph Kern, and Juan Carlos PerdomoIn Proceedings of the 42nd International Conference on Machine Learning (ICML), 2025🏆 Outstanding Paper Award at ICML 2025, Highlight Track at FORC 2025
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.
- Christoph Kern, Unai Fischer-Abaigar, Jonas Schweisthal, Dennis Frauen, Rayid Ghani, Stefan Feuerriegel, Mihaela Schaar, and Frauke KreuterNature Computational Science, 2025
Decision-making inherently involves cause–effect relationships that introduce causal challenges. We argue that reliable algorithms for decision-making need to build upon causal reasoning. Addressing these causal challenges requires explicit assumptions about the underlying causal structure to ensure identifiability and estimatability, which means that the computational methods must successfully align with decision-making objectives in real-world tasks.
2024
- Unai Fischer-Abaigar, Christoph Kern, Noam Barda, and Frauke KreuterGovernment Information Quarterly, 2024
AI-driven decision-making systems are becoming instrumental in the public sector, with applications spanning areas like criminal justice, social welfare, financial fraud detection, and public health. While these systems offer great potential benefits to institutional decision-making processes, such as improved efficiency and reliability, these systems face the challenge of aligning machine learning (ML) models with the complex realities of public sector decision-making. In this paper, we examine five key challenges where misalignment can occur, including distribution shifts, label bias, the influence of past decision-making on the data side, as well as competing objectives and human-in-the-loop on the model output side. Our findings suggest that standard ML methods often rely on assumptions that do not fully account for these complexities, potentially leading to unreliable and harmful predictions. To address this, we propose a shift in modeling efforts from focusing solely on predictive accuracy to improving decision-making outcomes. We offer guidance for selecting appropriate modeling frameworks, including counterfactual prediction and policy learning, by considering how the model estimand connects to the decision-maker’s utility. Additionally, we outline technical methods that address specific challenges within each modeling approach. Finally, we argue for the importance of external input from domain experts and stakeholders to ensure that model assumptions and design choices align with real-world policy objectives, taking a step towards harmonizing AI and public sector objectives.
- Unai Fischer-Abaigar, Christoph Kern, and Frauke KreuterIn Workshop on Humans, Algorithmic Decision-Making and Society, ICML 2024, 2024
Automated decision-making (ADM) systems are being deployed across a diverse range of critical problem areas such as social welfare and healthcare. Recent work highlights the importance of causal ML models in ADM systems, but implementing them in complex social environments poses significant challenges. Research on how these challenges impact the performance in specific downstream decision-making tasks is limited. Addressing this gap, we make use of a comprehensive real-world dataset of jobseekers to illustrate how the performance of a single CATE model can vary significantly across different decision-making scenarios and highlight the differential influence of challenges such as distribution shifts on predictions and allocations.
2022
- Christian Götzl, Selina Hiller, Christian Rauschenberg, Anita Schick, Janik Fechtelpeter, Unai Fischer-Abaigar, Georgia Koppe, Daniel Durstewitz, Ulrich Reininghaus, and Silvia KrummChild and Adolescent Psychiatry and Mental Health, 2022
- Unai Fischer-Abaigar2022Master’s thesis supervised by Daniel Durstewitz
Ordinal data, crucial for many scientific disciplines, consists of a discrete set of labels for which a natural ordering but no specified distance measure exists. In practice, this pecu- liarity of ordinal data is oftentimes overlooked, with many models making the simplified assumption that it can be interpreted as either metric or categorical. The rise of digital tech- nologies allows the collection of ever larger data sets, facilitating the use of more powerful and expressive machine learning architectures. This thesis proposes and evaluates a deep probabilistic latent model for forecasting ordinal time series by integrating an ordered-logit model into a sequential variational autoencoder framework. The model is developed in the context of a multi-university research initiative (Living lab AI4U) with the goal to predict in- dividuals’ emotional trajectories using time series data collected from questionnaires on their smartphones to suggest personalized mental health interventions. The model is evaluated using empirical data collected during a psychiatric study and benchmark data matching this data is created to test the model’s theoretical limitations. Hierarchical parameter estimation is implemented to deal with sparse and short time series. The findings identify future av- enues for dealing with irregular time series, missing values and ways to integrate multimodal sensor data from smartphones.