{"paper":{"title":"Causal Multi-Task Demand Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A meta-learning framework identifies causal demand parameters across tasks by conditioning on all prices while masking two outcomes for supervision.","cross_cats":["econ.EM","stat.ML"],"primary_cat":"cs.LG","authors_text":"Varun Gupta, Vijay Kamble","submitted_at":"2026-02-10T16:58:50Z","abstract_excerpt":"We study a canonical multi-task demand-learning problem motivated by retail pricing, where a firm seeks to estimate heterogeneous linear price-response functions across multiple decision contexts. Each context is described by rich covariates but exhibits limited price variation, motivating transfer learning across tasks. A central challenge in leveraging cross-task transfer is endogeneity: prices may be arbitrarily correlated with unobserved task-level demand determinants across tasks.\n  We propose a new meta-learning framework that identifies the conditional mean of task-specific causal deman"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We propose a new meta-learning framework that identifies the conditional mean of task-specific causal demand parameters given a subset of task-specific observables despite such confounding, assuming that each task contains at least two distinct locally exogenous price points.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"Each task contains at least two distinct locally exogenous price points, and the proposed information design (conditioning on all prices while masking two outcomes) is maximally uniformly valid.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A meta-learning method identifies the conditional mean of task-specific causal demand parameters by conditioning on all prices while masking two demand outcomes, assuming at least two locally exogenous prices per task.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A meta-learning framework identifies causal demand parameters across tasks by conditioning on all prices while masking two outcomes for supervision.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e0793d045082de55bb4c6ffdb61de200acd360f6727ee08fbbf87855e8ff0e36"},"source":{"id":"2602.09969","kind":"arxiv","version":2},"verdict":{"id":"2c6302bc-70e3-4381-89d9-25324db4a8ee","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T02:19:33.695599Z","strongest_claim":"We propose a new meta-learning framework that identifies the conditional mean of task-specific causal demand parameters given a subset of task-specific observables despite such confounding, assuming that each task contains at least two distinct locally exogenous price points.","one_line_summary":"A meta-learning method identifies the conditional mean of task-specific causal demand parameters by conditioning on all prices while masking two demand outcomes, assuming at least two locally exogenous prices per task.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"Each task contains at least two distinct locally exogenous price points, and the proposed information design (conditioning on all prices while masking two outcomes) is maximally uniformly valid.","pith_extraction_headline":"A meta-learning framework identifies causal demand parameters across tasks by conditioning on all prices while masking two outcomes for supervision."},"references":{"count":13,"sample":[{"doi":"10.1109/ijcnn.1991.155621","year":1991,"title":"Steven Berry, James Levinsohn, and Ariel Pakes","work_id":"71f1db2f-1bf4-45f8-904f-452448fb6d61","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1201/9781420057669","year":null,"title":"doi: 10.1201/9781420057669","work_id":"9c74a144-c8f5-4a79-a225-c28c252abb28","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.24432/c5bw33","year":null,"title":"Online Retail.https://doi.org/10.24432/C5BW33","work_id":"21f51930-d101-4d32-9f46-d3bd65d9f826","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Applied Causal Inference Powered by ML and AI","work_id":"5282d19c-4a82-45ab-8fea-522709da8b09","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1017/cbo9780511761362","year":null,"title":"doi: 10.1017/CBO9780511761362","work_id":"b81fe8aa-3533-46ff-958c-cce877cbb571","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":13,"snapshot_sha256":"fbb7715b68c395b70e22da89209a55558148c8e142b958e3f881c5091ec1e8b0","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"721c58fd7510ba97010f2076d311a6a0783b428c0e1ad13641cfc4dd47a28dbc"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}