# Fatigue-Aware Learning to Defer via Constrained Optimisation

arXiv: https://arxiv.org/abs/2604.00904 · HTML: https://mlsanity.com/paper/2604.00904/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 66 references that are in the map.
- First posted on arXiv 1 Apr 2026.

## Builds on (most cited first, 26 of 26)

- [Efficiently Modeling Long Sequences with Structured State Spaces](https://mlsanity.com/paper/2111.00396/) (2021)
- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [Simplified State Space Layers for Sequence Modeling](https://mlsanity.com/paper/2208.04933/) (2022)
- [Hybrid LLM: Cost-Efficient and Quality-Aware Query Routing](https://mlsanity.com/paper/2404.14618/) (2024)
- [Consistent Estimators for Learning to Defer to an Expert](https://mlsanity.com/paper/2006.01862/) (2020)
- [Stabilizing Transformers for Reinforcement Learning](https://mlsanity.com/paper/1910.06764/) (2019)
- [Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotations](https://mlsanity.com/paper/2110.12088/) (2021)
- [Predict Responsibly: Improving Fairness and Accuracy by Learning to Defer](https://mlsanity.com/paper/1711.06664/) (2017)
- [Structured State Space Models for In-Context Reinforcement Learning](https://mlsanity.com/paper/2303.03982/) (2023)
- [Gradients are Not All You Need](https://mlsanity.com/paper/2111.05803/) (2021)
- [Calibrated Learning to Defer with One-vs-All Classifiers](https://mlsanity.com/paper/2202.03673/) (2022)
- [Do We Train on Test Data? Purging CIFAR of Near-Duplicates](https://mlsanity.com/paper/1902.00423/) (2019)
- [POPGym: Benchmarking Partially Observable Reinforcement Learning](https://mlsanity.com/paper/2303.01859/) (2023)
- [Learning to Defer to Multiple Experts: Consistent Surrogate Losses, Confidence Calibration, and Conformal Ensembles](https://mlsanity.com/paper/2210.16955/) (2022)
- [Hard Sample Aware Noise Robust Learning for Histopathology Image Classification](https://mlsanity.com/paper/2112.03694/) (2021)
- [OPTIMAM Mammography Image Database: a large scale resource of mammography images and clinical data](https://mlsanity.com/paper/2004.04742/) (2020)
- [Regression with Multi-Expert Deferral](https://mlsanity.com/paper/2403.19494/) (2024)
- [Learning to Defer to a Population: A Meta-Learning Approach](https://mlsanity.com/paper/2403.02683/) (2024)
- [Decision S4: Efficient Sequence-Based RL via State Spaces Layers](https://mlsanity.com/paper/2306.05167/) (2023)
- [In Defense of Softmax Parametrization for Calibrated and Consistent Learning to Defer](https://mlsanity.com/paper/2311.01106/) (2023)
- [OvercookedV2: Rethinking Overcooked for Zero-Shot Coordination](https://mlsanity.com/paper/2503.17821/) (2025)
- [Cross-environment Cooperation Enables Zero-shot Multi-agent Coordination](https://mlsanity.com/paper/2504.12714/) (2025)
- [Learning-to-defer for sequential medical decision-making under uncertainty](https://mlsanity.com/paper/2109.06312/) (2021)
- [A data-centric approach for improving ambiguous labels with combined semi-supervised classification and clustering](https://mlsanity.com/paper/2106.16209/) (2021)
- [Learning to Complement and to Defer to Multiple Users](https://mlsanity.com/paper/2407.07003/) (2024)
- [Coverage-Constrained Human-AI Cooperation with Multiple Experts](https://mlsanity.com/paper/2411.11976/) (2024)

## Built on by (most cited first, 3 of 3)

- [Oversight Has a Capacity: Calibrating Agent Guards to a Subjective, Fatiguing Human](https://mlsanity.com/paper/2606.08919/) (2026)
- [Learning to Assign Prediction Tasks to Agents with Capacity Constraints](https://mlsanity.com/paper/2605.27999/) (2026)
- [PixSim: a calibrated open-source simulator of instant-payment fraud, recovery and interdiction under analyst capacity constraints](https://mlsanity.com/paper/2609.30684/) (2026)

## Abstract (arXiv)

Learning to defer (L2D) enables human-AI cooperation by deciding when an AI system should act autonomously or defer to a human expert. Existing L2D methods, however, assume static human performance, contradicting well-established findings on fatigue-induced degradation. We propose Fatigue-Aware Learning to Defer via Constrained Optimisation (FALCON), which explicitly models workload-varying human performance using psychologically grounded fatigue curves. FALCON formulates L2D as a Constrained Markov Decision Process (CMDP) whose state includes both task features and cumulative human workload, and optimises accuracy under human-AI cooperation budgets via PPO-Lagrangian training. We further introduce FA-L2D, a benchmark that systematically varies fatigue dynamics from near-static to rapidly degrading regimes. Experiments across multiple datasets show that FALCON consistently outperforms state-of-the-art L2D methods across coverage levels, generalises zero-shot to unseen experts with different fatigue patterns, and demonstrates the advantage of adaptive human-AI collaboration over AI-only or human-only decision-making when coverage lies strictly between 0 and 1.