Towards Value-Constrained Credit Assignment in Fully Delegated AI Cooperatives
2026-06-26 • Machine Learning
Machine LearningArtificial IntelligenceDistributed, Parallel, and Cluster ComputingMultiagent Systems
AI summaryⓘ
The authors propose a new method for fairly sharing rewards in AI projects where humans delegate work to agents that provide data and update models. Their idea is to only give credit for updates that align with each person's values using a process called value-conditioned gradient filtering. They use a decentralized learning system named traversal learning, which avoids quality loss seen in other methods and helps track who contributed what more precisely. This approach improves on existing data valuation and federated learning techniques by better respecting individual preferences and contributions.
reward allocationfederated learninggradient filteringtraversal learningdata valuationdecentralized backpropagationmodel updatesmarginal contributionpluralistic alignmentpersonalized learning
Authors
Young Yoon, Jimin Kim, Soyeon Park
Abstract
We propose a framework for reward allocation in fully delegated AI cooperatives where humans are represented by agents that contribute data and participate in model updates under heterogeneous value constraints. The key idea is to credit only those updates that remain admissible after screening them against each principal's value profile. We formulate value-conditioned gradient filtering, online marginal contribution signals, and cumulative revenue settlement within a traversal learning (TL) substrate. TL is especially attractive here because it performs decentralized backpropagation without the quality loss associated with aggregation-centric distributed learning and, we argue, offers a finer attribution substrate than FedAvg-style federated learning by preserving explicit traversal and gradient paths. The framework is positioned against data valuation, federated contribution estimation, personalized federated learning, and pluralistic alignment.