{"paper":{"title":"Ready from Day 1: Population-Aware Coordination for Large-Scale Constrained Multi-Agent Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Population-aware learned maps let planners coordinate large multi-agent systems across changing compositions without retraining.","cross_cats":["cs.LG"],"primary_cat":"cs.MA","authors_text":"Alvaro Maggiar, Angel Wang, Carson Eisenach, Dean Foster, Dominique Perrault-Joncas","submitted_at":"2026-05-12T16:57:24Z","abstract_excerpt":"In large-scale multi-agent systems with shared resource constraints, an upstream planner must iteratively evaluate candidate resource plans -- assessing feasibility, aggregate response, and marginal cost -- before committing to one. Lagrangian relaxation separates local decisions through a broadcast cost signal, but the planner still needs the cost-to-utilization response map to explore plan space, and this map depends on population composition that changes across planning cycles. We propose \\emph{population-aware coordination interfaces}: learned primal and dual maps, conditioned on compact p"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"In a supply-chain capacity-control case study, population-aware interfaces reduce forecast error by 16--19% and capacity violations by 20--51% relative to population-unaware baselines under composition shift; 20K-agent cohorts support accurate coordination of 500K-agent populations; and simulator-trained primal maps achieve 11.1% MAPE on real observations versus 13--24% for baselines.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That compact population summaries encode sufficient response-relevant structure for the learned primal and dual maps to remain reliable across evolving populations without per-cycle retraining.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Population-conditioned learned primal and dual maps support reliable coordination of large multi-agent systems under composition shifts without per-cycle retraining, cutting forecast error 16-19% and violations 20-51% in supply-chain tests.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Population-aware learned maps let planners coordinate large multi-agent systems across changing compositions without retraining.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"3f0a69b4635a59c515f04539b7111512c134118a22b0b5752b6b22c79388f12e"},"source":{"id":"2605.13900","kind":"arxiv","version":1},"verdict":{"id":"7955b3a3-4bbf-4317-b95e-a51862f8a10f","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T04:58:48.008522Z","strongest_claim":"In a supply-chain capacity-control case study, population-aware interfaces reduce forecast error by 16--19% and capacity violations by 20--51% relative to population-unaware baselines under composition shift; 20K-agent cohorts support accurate coordination of 500K-agent populations; and simulator-trained primal maps achieve 11.1% MAPE on real observations versus 13--24% for baselines.","one_line_summary":"Population-conditioned learned primal and dual maps support reliable coordination of large multi-agent systems under composition shifts without per-cycle retraining, cutting forecast error 16-19% and violations 20-51% in supply-chain tests.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That compact population summaries encode sufficient response-relevant structure for the learned primal and dual maps to remain reliable across evolving populations without per-cycle retraining.","pith_extraction_headline":"Population-aware learned maps let planners coordinate large multi-agent systems across changing compositions without retraining."},"references":{"count":45,"sample":[{"doi":"","year":2003,"title":"CACHON, G. P. (2003). Supply chain coordination with contracts. InHandbooks in Operations Research and Management Science, vol. 11. Elsevier, 227–339","work_id":"ee151e89-05c7-4c24-a4b9-abfb2bef5d4d","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1999,"title":"FEDERGRUEN, A. and ZIPKIN, P. H. (1999). Coordination mechanisms for a distribution system with one supplier and multiple retailers.Management science451493–1507","work_id":"84747aec-ba7e-47a5-917e-e8e9e08d00d7","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2011,"title":"BOYD, S., PARIKH, N., CHU, E., PELEATO, B. and ECKSTEIN, J. (2011). Distributed opti- mization and statistical learning via the alternating direction method of multipliers.Foundations and Trends in Ma","work_id":"16f5a3ce-3379-4759-9ca9-63d6f32bd3a9","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1981,"title":"FISHER, M. L. (1981). The lagrangian relaxation method for solving integer programming problems.Management science271–18","work_id":"eb40f41a-6d11-485d-9104-4de30a80e625","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"LOWE, R., WU, Y., TAMAR, A., HARB, J., ABBEEL, P. and MORDATCH, I. (2017). Multi- agent actor-critic for mixed cooperative-competitive environments. InAdvances in Neural Information Processing Systems","work_id":"3ad30ad7-ff6e-452c-a00f-db72c9598866","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":45,"snapshot_sha256":"c2e1daaa13517e08411150ee243bb35454a45f9793a23cee8a244fe696a21009","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}