mlsanityData to 14 Oct 2026

A Two-Stage Learning-to-Defer Approach for Multi-Task Learning

arXiv 2410.15729 ↗First posted 21 Oct 2024stat.ML cs.HC cs.LGcited by 17

Not on the map. Why?

Abstract (arXiv)

The Two-Stage Learning-to-Defer (L2D) framework has been extensively studied for classification and, more recently, regression tasks. However, many real-world applications require solving both tasks jointly in a multi-task setting. We introduce a novel Two-Stage L2D framework for multi-task learning that integrates classification and regression through a unified deferral mechanism. Our method leverages a two-stage surrogate loss family, which we prove to be both Bayes-consistent and (G, R)-consistent, ensuring convergence to the Bayes-optimal rejector. We derive explicit consistency bounds tied to the cross-entropy surrogate and the L1-norm of agent-specific costs, and extend minimizability gap analysis to the multi-expert two-stage regime. We also make explicit how shared representation learning -- commonly used in multi-task models -- affects these consistency guarantees. Experiments on object detection and electronic health record analysis demonstrate the effectiveness of our approach and highlight the limitations of existing L2D methods in multi-task scenarios.

Key facts. Cited by 17 papers in the map (data through 14 Oct 2026). Builds on 32 of its 73 references that are in the map. First posted on arXiv 21 Oct 2024.

Builds on

32 of 73 references.

References · 32

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  3. 2020End-to-End Object Detection with Transformerscited by 7,414
  4. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,852
  5. 2015Fast R-CNNcited by 2,933
  6. 2017Selective Classification for Deep Neural Networkscited by 852
  7. 2017Robust Loss Functions under Label Noise for Deep Neural Networkscited by 443
  8. 2020Consistent Estimators for Learning to Defer to an Expertcited by 271
  9. 2017Predict Responsibly: Improving Fairness and Accuracy by Learning to Defercited by 233
  10. 2023Cross-Entropy Loss Functions: Theoretical Analysis and Applicationscited by 210
  11. 2021Differentiable Learning Under Triagecited by 76
  12. 2021Towards Unbiased and Accurate Deferral to Multiple Expertscited by 72
20 more
  1. 2023Who Should Predict? Exact Algorithms For Learning to Defer to Humanscited by 70
  2. 2022Calibrated Learning to Defer with One-vs-All Classifierscited by 69
  3. 2022Learning to Defer to Multiple Experts: Consistent Surrogate Losses, Confidence Calibration, and Conformal Ensemblescited by 60
  4. 2022Sample Efficient Learning of Predictors that Complement Humanscited by 46
  5. 2021Combining Human Predictions with Model Probabilities via Confusion Matrices and Calibrationcited by 40
  6. 2022Forming Effective Human-AI Teams: Building Machine Learning Models that Complement the Capabilities of Multiple Expertscited by 40
  7. 2023Principled Approaches for Learning to Defer with Multiple Expertscited by 38
  8. 2023Predictor-Rejector Multi-Class Abstention: Theoretical Analysis and Algorithmscited by 36
  9. 2023Theoretically Grounded Loss Functions and Algorithms for Score-Based Multi-Class Abstentioncited by 32
  10. 2024Regression with Multi-Expert Deferralcited by 29
  11. 2024Learning to Defer to a Population: A Meta-Learning Approachcited by 27
  12. 2023In Defense of Softmax Parametrization for Calibrated and Consistent Learning to Defercited by 26
  13. 2024Realizable H-Consistent and Bayes-Consistent Loss Functions for Learning to Defercited by 26
  14. 2024A Universal Growth Rate for Learning with Smooth Surrogate Lossescited by 24
  15. 2024Cardinality-Aware Set Prediction and Top-k Classificationcited by 23
  16. 2024Multi-Label Learning with Stronger Consistency Guaranteescited by 22
  17. 2024H-Consistency Guarantees for Regressioncited by 21
  18. 2024Enhanced H-Consistency Boundscited by 18
  19. 2022Counterfactual Inference of Second Opinionscited by 14
  20. 2024A Causal Framework for Evaluating Deferring Systemscited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 9; 2026: 8.

17 papers.

Sort:

  1. 2025Adversarial Robustness in Two-Stage Learning-to-Defer: Algorithms and Guaranteescited by 16
  2. 2025Mastering Multiple-Expert Routing: Realizable H-Consistency and Strong Guarantees for Learning to Defercited by 16
  3. 2025Balancing the Scales: A Theoretical and Algorithmic Framework for Learning from Imbalanced Datacited by 15
  4. 2025Budgeted Multiple-Expert Deferralcited by 13
  5. 2025Why Ask One When You Can Ask k? Learning-to-Defer to the Top-k Expertscited by 7
  6. 2025Principled Algorithms for Optimizing Generalized Metrics in Binary Classificationcited by 6
  7. 2026Learning-to-Defer in Non-Stationary Time Series via Switching State-Space Modelscited by 6
  8. 2025Adversarial Robustness in One-Stage Learning-to-Defercited by 5
  9. 2025Identity-Free Deferral For Unseen Expertscited by 4
  10. 2026When More Experts Hurt: Underfitting in Multi-Expert Learning to Defercited by 3
  11. 2026Consistent Learning-to-Defer with Expert-Conditional Advicecited by 3
  12. 2026Optimized Deferral for Imbalanced Settingscited by 2
5 more
  1. 2026Linear-Core Surrogates: Smooth Loss Functions with Linear Rates for Classification and Structured Predictioncited by 2
  2. 2026Coherent Hierarchical Multi-Label Learning to Defer for Medical Imagingcited by 1
  3. 2026Online Learning-to-Defer with Varying Expertscited by 1
  4. 2025Uncertainty-Guided Expert-AI Collaboration for Efficient Soil Horizon Annotationcited by 0
  5. 2026Should I stay or should I show? Learning to selectively disclose informationcited by 0