Optimized Deferral for Imbalanced Settings

• URL: http://arxiv.org/abs/2604.27723v1
• Date: Thu, 30 Apr 2026 11:15:37 GMT
• Title: Optimized Deferral for Imbalanced Settings
• Abstract summary: Learning algorithms can be significantly improved by routing complex or uncertain inputs to specialized experts.<n>This approach, known as learning to defer, is essential in domains like natural language generation, medical diagnosis, and computer vision.<n>We present a comprehensive study of two-stage learning to defer in expert imbalance settings.
• Score: 52.48138985215643
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Learning algorithms can be significantly improved by routing complex or uncertain inputs to specialized experts, balancing accuracy with computational cost. This approach, known as learning to defer, is essential in domains like natural language generation, medical diagnosis, and computer vision, where an effective deferral can reduce errors at low extra resource consumption. However, the two-stage learning to defer setting, which leverages existing predictors such as a collection of LLMs or other classifiers, often faces challenges due to an expert imbalance problem. This imbalance can lead to suboptimal performance, with deferral algorithms favoring the majority expert. We present a comprehensive study of two-stage learning to defer in expert imbalance settings. We cast the deferral loss optimization as a novel cost-sensitive learning problem over the input-expert domain. We derive new margin-based loss functions and guarantees tailored to this setting, and develop novel algorithms for cost-sensitive learning. Leveraging these results, we design principled deferral algorithms, MILD (Margin-based Imbalanced Learning to Defer), specifically suited for expert imbalance settings. Extensive experiments demonstrate the effectiveness of our approach, showing clear improvements over existing baselines on both image classification and real-world Large Language Model (LLM) routing tasks.

Related papers

• A Replicate-and-Quantize Strategy for Plug-and-Play Load Balancing of Sparse Mixture-of-Experts LLMs [64.8510381475827]
  Sparse Mixture-of-Experts (SMoE) architectures are increasingly used to scale large language models efficiently.<n>SMoE models often suffer from severe load imbalance across experts, where a small subset of experts receives most tokens while others are underutilized.<n>We present a systematic analysis of expert routing during inference and identify three findings: (i) load imbalance persists and worsens with larger batch sizes, (ii) selection frequency does not reliably reflect expert importance, and (iii) overall expert workload and importance can be estimated using a small calibration set.
  arXiv  Detail & Related papers  (2026-02-23T15:11:16Z)
• Budgeted Multiple-Expert Deferral [38.13580998392063]
  Training procedures for deferral algorithms typically require querying all experts for every training instance.<n>We introduce the budgeted deferral framework, which aims to train effective deferral algorithms while minimizing expert query costs during training.<n>We propose new algorithms for both two-stage and single-stage multiple-expert deferral settings that selectively query only a subset of experts per training example.
  arXiv  Detail & Related papers  (2025-10-30T17:08:52Z)
• OptiMind: Teaching LLMs to Think Like Optimization Experts [14.155402554740997]
  We systematically integrate optimization expertise to improve formulation accuracy for mixed-integer linear programming.<n>Our approach first cleans training data through class-based error analysis.<n>We then develop multi-turn inference strategies that guide LLMs with class-specific error summaries and solver feedback.
  arXiv  Detail & Related papers  (2025-09-26T22:23:12Z)
• ALScope: A Unified Toolkit for Deep Active Learning [5.2705718569212285]
  Deep Active Learning (DAL) reduces annotation costs by selecting the most informative unlabeled samples during training.<n>We present a new DAL platform ALScope for classification tasks, integrating 10 datasets from computer vision (CV) and natural language processing (NLP)<n>This platform supports flexible configuration of key experimental factors, ranging from algorithm and dataset choices to task-specific factors like out-of-distribution (OOD) sample ratio.
  arXiv  Detail & Related papers  (2025-08-06T23:39:46Z)
• Addressing Imbalanced Domain-Incremental Learning through Dual-Balance Collaborative Experts [59.615381619866284]
  Domain-Incremental Learning (DIL) focuses on continual learning in non-stationary environments.<n>DIL faces two critical challenges in the context of imbalanced data: intra-domain class imbalance and cross-domain class distribution shifts.<n>We introduce the Dual-Balance Collaborative Experts (DCE) framework to overcome these challenges.
  arXiv  Detail & Related papers  (2025-07-09T17:57:07Z)
• Learning to Optimize for Reinforcement Learning [58.01132862590378]
  Reinforcement learning (RL) is essentially different from supervised learning, and in practice, these learneds do not work well even in simple RL tasks. Agent-gradient distribution is non-independent and identically distributed, leading to inefficient meta-training. We show that, although only trained in toy tasks, our learned can generalize unseen complex tasks in Brax.
  arXiv  Detail & Related papers  (2023-02-03T00:11:02Z)
• Learning Algorithms for Intelligent Agents and Mechanisms [4.251500966181852]
  In this thesis, we research learning algorithms for optimal decision making in two different contexts, Reinforcement Learning in Part I and Auction Design in Part II. In Chapter 2, inspired by statistical physics, we develop a novel approach to Reinforcement Learning (RL) that not only learns optimal policies with enhanced desirable properties but also sheds new light on maximum entropy RL. In Chapter 3, we tackle the generalization problem in RL using a Bayesian perspective. We show that imperfect knowledge of the environments dynamics effectively turn a fully-observed Markov Decision Process (MDP) into a Partially Observed MDP (POMD
  arXiv  Detail & Related papers  (2022-10-06T03:12:43Z)
• Improved Algorithms for Neural Active Learning [74.89097665112621]
  We improve the theoretical and empirical performance of neural-network(NN)-based active learning algorithms for the non-parametric streaming setting. We introduce two regret metrics by minimizing the population loss that are more suitable in active learning than the one used in state-of-the-art (SOTA) related work.
  arXiv  Detail & Related papers  (2022-10-02T05:03:38Z)
• Reparameterized Variational Divergence Minimization for Stable Imitation [57.06909373038396]
  We study the extent to which variations in the choice of probabilistic divergence may yield more performant ILO algorithms. We contribute a re parameterization trick for adversarial imitation learning to alleviate the challenges of the promising $f$-divergence minimization framework. Empirically, we demonstrate that our design choices allow for ILO algorithms that outperform baseline approaches and more closely match expert performance in low-dimensional continuous-control tasks.
  arXiv  Detail & Related papers  (2020-06-18T19:04:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.