paper
Towards Value-Constrained Credit Assignment in Fully Delegated AI Cooperatives
paperactiveprovisional
towards-value-constrained-credit-assignment-in-fully-delegated-ai-cooperatives-36fe5813·1 events·first seen 45h agoAliases: Towards Value-Constrained Credit Assignment in Fully Delegated AI Cooperatives
Co-occurring entities
More like this (12)
token credit assignmentVals AIVoluntary AI Safety CommitmentsVals AI Finance Agent BenchmarkAI-driven constraint reasoningEfficient and Sound Probabilistic Verification for AI AgentsCan Trustless Agents Be Trusted? An Empirical Study of the ERC-8004 Decentralized AI Agent EcosystemTrustworthy AImulti-agent cooperative frameworkDemocratic Inputs to AIDemocratic ICAI: Debating Our Way to Steering Principles from PreferencesCohere for AI
Recent events (1)
Framework for value-constrained credit assignment in fully delegated AI cooperatives
A new arXiv preprint proposes a framework for reward allocation in AI cooperatives where human principals are represented by agents contributing data and model updates under heterogeneous value constraints. The approach introduces value-conditioned gradient filtering and online marginal contribution signals within a 'traversal learning' (TL) substrate, which the authors argue preserves explicit gradient paths and enables finer attribution than FedAvg-style federated learning. The work positions itself against data valuation, federated contribution estimation, personalized federated learning, and pluralistic alignment research.