Preference Elicitation for Policy Optimization and Application to Aligning Heart Transplantation with Human Values

Machine learning predictors have become essential tools for guiding automated decision making. However, a major misalignment persists: predictive models are typically optimized in terms of standard statistical metrics in isolation from the algorithmic tasks they inform. We highlight this incongruity in the high-stakes domain of organ allocation by demonstrating that any algorithm relying on (even highly accurate) survival predictors optimized for standard metrics -- such as the Concordance index (C-index) -- can yield arbitrarily poor outcomes when used for allocation, failing to guarantee utility better than a uniform random selection. To bridge the gap between survival analysis and policy optimization, we introduce a decision-focused learning approach based on optimizing normalized discounted cumulative gain (NDCG), a mainstay metric in information retrieval. We establish the utility of NDCG in survival analysis by proving that it translates to guarantees on the performance of allocation. Empirically, we propose a bootstrapping approach to optimize the NDCG of existing survival models. Unlike prior work, we also address the challenge of right censorship when evaluating ranking. On historical heart transplant data from the US, our method dramatically boosts the NDCG of baseline models by 50-100%, which translates to tens of thousands of additional life years gained annually when deployed for transplant allocation. We anticipate that our framework will find broader applications in decision making with predictions.

Pluralistic alignment needs to accommodate diverse preferences under limited human feedback. Using AI simulators to label preference comparisons reduces annotation costs, but variation in simulation accuracy across individuals can leave some people's preferences poorly represented in training. We study how to allocate human feedback across people to construct training data for pluralistic alignment, and how this allocation affects the trained policy. To make this allocation explicit and controllable, we define a social welfare objective over individual supervision utilities that measure label quality. Each query's marginal welfare gain depends on its expected supervision improvement, the recipient's current utility, and application-designated priority. Welfare-Aware Preference Acquisition (WAPA) uses auxiliary reference models to estimate query gains and individual utilities, then selects queries by estimated marginal welfare gain while updating recipient utilities. We prove exact optimality of the acquisition algorithm for the estimated welfare objective under a fixed query budget and model predictions. Experiments on six Roleplay collections and four OpinionQA regions demonstrate improved supervision and controllable allocation tradeoffs. Policy training on four Roleplay collections improves preference prediction for both seen and unseen individuals. When prioritizing a designated group and poorly represented individuals, acquiring human labels for 20% of comparisons recovers 75-86% of the average predictive utility gap between policies trained entirely on AI-generated labels and entirely on ground-truth labels.

When an AI algorithm makes decisions that affect more than one person, aligning it becomes a problem of social choice: how should people's divergent preferences about system behavior be reconciled and aggregated into a single coherent model? The standard approach to aligning frontier AI models$\unicode{x2013}$reinforcement learning from human feedback$\unicode{x2013}$largely sidesteps this question and has poor social choice guarantees. However, it remains unclear what alternative should replace it. We show that, by focusing directly on an algorithm's welfare consequences, the alignment problem can be reformulated as linear optimization over a convex impact space, which makes it amenable to the standard toolkit of welfare economics and mechanism design. This reformulation clarifies how alignment protocols translate into welfare consequences and, conversely, how a social planner's desired constraints on welfare consequences can be translated back into alignment protocols. We apply this transformation to show that voting-by-issues and random-dictatorship mechanisms are strategyproof and unanimous. Demonstrating the reverse direction, we also apply the impact representation to derive a family of alignment protocols that maximize utilitarian social welfare subject to various social desiderata, such as bounds on individual or group harm. We illustrate the welfare implications of these alignment protocols empirically using real human preferences over kidney allocation, charitable food distribution, LLM responses, and trolley problems.