{"paper":{"title":"TG-DIN: Theory-Guided Demand Inference Network for Generalizable QoS Measurement and Prediction","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"A neural network infers latent user demand from QoS measurements by embedding scheduling and queuing rules as a differentiable theory layer.","cross_cats":[],"primary_cat":"cs.NI","authors_text":"Feng Ye, Fuliang Yang","submitted_at":"2026-05-15T02:43:35Z","abstract_excerpt":"In this paper, we introduce TG-DIN, a theory-guided demand inference network that infers latent user demand from observable network quality-of-service (QoS) measurements. Rather than directly predicting QoS outcomes using black-box models, TG-DIN explicitly models latent demand as an intermediate variable and links it to observable behavior through a differentiable theory layer grounded in scheduling and queuing principles. This design yields an interpretable, mechanism-consistent representation of user demand that is directly applicable to downstream tasks such as congestion diagnosis, resour"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"TG-DIN explicitly models latent demand as an intermediate variable and links it to observable behavior through a differentiable theory layer grounded in scheduling and queuing principles. This design yields an interpretable, mechanism-consistent representation of user demand that is directly applicable to downstream tasks such as congestion diagnosis, resource allocation, capacity planning, and policy evaluation.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The differentiable theory layer grounded in scheduling and queuing principles accurately captures the real mechanisms that connect latent demand to observable QoS measurements in both synthetic and real network settings.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A neural network with a theory-guided differentiable layer infers hidden demand from QoS data for improved generalization across network conditions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A neural network infers latent user demand from QoS measurements by embedding scheduling and queuing rules as a differentiable theory layer.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9f574ea45df3d2dc8be784fa1b4532c244a9ff93b658576b07a3eb86c24bc685"},"source":{"id":"2605.15550","kind":"arxiv","version":1},"verdict":{"id":"66211b20-5244-464b-ae19-ba4de4b85ade","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T20:00:27.183811Z","strongest_claim":"TG-DIN explicitly models latent demand as an intermediate variable and links it to observable behavior through a differentiable theory layer grounded in scheduling and queuing principles. This design yields an interpretable, mechanism-consistent representation of user demand that is directly applicable to downstream tasks such as congestion diagnosis, resource allocation, capacity planning, and policy evaluation.","one_line_summary":"A neural network with a theory-guided differentiable layer infers hidden demand from QoS data for improved generalization across network conditions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The differentiable theory layer grounded in scheduling and queuing principles accurately captures the real mechanisms that connect latent demand to observable QoS measurements in both synthetic and real network settings.","pith_extraction_headline":"A neural network infers latent user demand from QoS measurements by embedding scheduling and queuing rules as a differentiable theory layer."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15550/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T20:31:19.247999Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T20:11:05.136658Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T19:34:36.248440Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:41:56.095622Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"2432a756a570125f4e23bb311fa14961746913c2151d3f5b48e569a62389d17a"},"references":{"count":37,"sample":[{"doi":"","year":2024,"title":"Abd AlRhman AlQiam, Yuanjun Yao, Zhaodong Wang, Satyajeet Singh Ahuja, Ying Zhang, Sanjay G Rao, Bruno Ribeiro, and Mohit Tawarmalani. 2024. Transfer- able neural wan te for changing topologies. InPro","work_id":"047b0172-6f23-4e8a-8363-7a39909bbf87","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1145/3474369.3486864","year":2021,"title":"A Framework for Cluster and Classifier Evaluation in the Absence of Reference Labels","work_id":"6df709d4-7cfe-42f0-bb44-c109010b2812","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Ons Aouedi, Van An Le, Kandaraj Piamrat, and Yusheng Ji. 2025. Deep learning on network traffic prediction: Recent advances, analysis, and future directions. ACM computing surveys57, 6 (2025), 1–37","work_id":"12866cd4-9c5a-4f12-897f-4b8d31ef3b7e","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2013,"title":"Athula Balachandran, Vyas Sekar, Aditya Akella, Srinivasan Seshan, Ion Stoica, and Hui Zhang. 2013. Developing a predictive model of quality of experience for internet video.ACM SIGCOMM Computer Commu","work_id":"4d28ad23-ee1b-4dbd-991f-3b3ea779c65f","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2000,"title":"Giuseppe Bianchi. 2000. Performance analysis of the IEEE 802.11 distributed coordination function.IEEE Journal on Selected Areas in Communications18, 3 (2000), 535–547","work_id":"9626a8a5-b469-4bcd-b748-7163091ce394","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":37,"snapshot_sha256":"f782941b739f50e7d3621e770b1c526a08b156a662cc5a2bf84d1ac4c0440aa1","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"}