Methods and Open Problems in Differentiable Social Choice: Learning Mechanisms, Decisions, and Alignment

2026-02-03

AI Summary

This work addresses the implicit implementation of social choice in modern machine learning systems—such as those used in auctions, federated learning, and large model alignment—which currently lack explicit normative frameworks and learnable aggregation mechanisms. The paper introduces the first differentiable social choice paradigm, modeling voting rules, preference aggregation, and mechanism design as differentiable structures that incorporate axiomatic constraints from social choice theory. By enabling data-driven optimization, this framework facilitates fair and transparent collective decision-making. It further reveals novel manifestations of classical impossibility theorems in learning environments and demonstrates empirical efficacy across diverse settings, including auctions, budget allocation, and liquid democracy. The study concludes by proposing 36 open problems spanning machine learning, economics, and democratic theory, laying a foundational basis for aligned and trustworthy AI-driven decision systems.

Technology Category

• Game Theory and Economic Paradigms: Social Choice / Voting
• Multiagent Systems: Mechanism Design
• Machine Learning: Learning Preferences or Rankings

Application Category

• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Social choice is no longer a peripheral concern of political theory or economics-it has become a foundational component of modern machine learning systems. From auctions and resource allocation to federated learning, participatory governance, and the alignment of large language models, machine learning pipelines increasingly aggregate heterogeneous preferences, incentives, and judgments into collective decisions. In effect, many contemporary machine learning systems already implement social choice mechanisms, often implicitly and without explicit normative scrutiny. This Review surveys differentiable social choice: an emerging paradigm that formulates voting rules, mechanisms, and aggregation procedures as learnable, differentiable models optimized from data. We synthesize work across auctions, voting, budgeting, liquid democracy, decentralized aggregation, and inverse mechanism learning, showing how classical axioms and impossibility results reappear as objectives, constraints, and optimization trade-offs. We conclude by identifying 36 open problems defining a new research agenda at the intersection of machine learning, economics, and social choice theory.

Problem

• social choice
• machine learning
• preference aggregation
• collective decision-making
• normative scrutiny

Innovation

• differentiable social choice
• learnable mechanisms
• preference aggregation
• inverse mechanism learning
• collective decision-making