Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Scalable Curiosity-Driven Game-Theoretic Framework for Long-Tail Multi-Label Learning in Data Mining

Jing Yang, Keze Wang

TL;DR

Curiosity-Driven Game-Theoretic Multi-Label Learning (CD-GTMLL), a scalable cooperative framework that recasts long-tail MLC as a multi-player game that enhances model efficiency in resource-constrained environments but also paves the way for more adaptive learning in imbalanced data scenarios across industries like e-commerce and healthcare.

Abstract

The long-tail distribution, where a few head labels dominate while rare tail labels abound, poses a persistent challenge for large-scale Multi-Label Classification (MLC) in real-world data mining applications. Existing resampling and reweighting strategies often disrupt inter-label dependencies or require brittle hyperparameter tuning, especially as the label space expands to tens of thousands of labels. To address this issue, we propose Curiosity-Driven Game-Theoretic Multi-Label Learning (CD-GTMLL), a scalable cooperative framework that recasts long-tail MLC as a multi-player game - each sub-predictor ("player") specializes in a partition of the label space, collaborating to maximize global accuracy while pursuing intrinsic curiosity rewards based on tail label rarity and inter-player disagreement. This mechanism adaptively injects learning signals into under-represented tail labels without manual balancing or tuning. We further provide a theoretical analysis showing that our CD-GTMLL converges to a tail-aware equilibrium and formally links the optimization dynamics to improvements in the Rare-F1 metric. Extensive experiments across 7 benchmarks, including extreme multi-label classification datasets with 30,000+ labels, demonstrate that CD-GTMLL consistently surpasses state-of-the-art methods, with gains up to +1.6% P@3 on Wiki10-31K. Ablation studies further confirm the contributions of both game-theoretic cooperation and curiosity-driven exploration to robust tail performance. By integrating game theory with curiosity mechanisms, CD-GTMLL not only enhances model efficiency in resource-constrained environments but also paves the way for more adaptive learning in imbalanced data scenarios across industries like e-commerce and healthcare.

A Scalable Curiosity-Driven Game-Theoretic Framework for Long-Tail Multi-Label Learning in Data Mining

TL;DR

Curiosity-Driven Game-Theoretic Multi-Label Learning (CD-GTMLL), a scalable cooperative framework that recasts long-tail MLC as a multi-player game that enhances model efficiency in resource-constrained environments but also paves the way for more adaptive learning in imbalanced data scenarios across industries like e-commerce and healthcare.

Abstract

The long-tail distribution, where a few head labels dominate while rare tail labels abound, poses a persistent challenge for large-scale Multi-Label Classification (MLC) in real-world data mining applications. Existing resampling and reweighting strategies often disrupt inter-label dependencies or require brittle hyperparameter tuning, especially as the label space expands to tens of thousands of labels. To address this issue, we propose Curiosity-Driven Game-Theoretic Multi-Label Learning (CD-GTMLL), a scalable cooperative framework that recasts long-tail MLC as a multi-player game - each sub-predictor ("player") specializes in a partition of the label space, collaborating to maximize global accuracy while pursuing intrinsic curiosity rewards based on tail label rarity and inter-player disagreement. This mechanism adaptively injects learning signals into under-represented tail labels without manual balancing or tuning. We further provide a theoretical analysis showing that our CD-GTMLL converges to a tail-aware equilibrium and formally links the optimization dynamics to improvements in the Rare-F1 metric. Extensive experiments across 7 benchmarks, including extreme multi-label classification datasets with 30,000+ labels, demonstrate that CD-GTMLL consistently surpasses state-of-the-art methods, with gains up to +1.6% P@3 on Wiki10-31K. Ablation studies further confirm the contributions of both game-theoretic cooperation and curiosity-driven exploration to robust tail performance. By integrating game theory with curiosity mechanisms, CD-GTMLL not only enhances model efficiency in resource-constrained environments but also paves the way for more adaptive learning in imbalanced data scenarios across industries like e-commerce and healthcare.
Paper Structure (67 sections, 5 theorems, 14 equations, 4 figures, 5 tables, 2 algorithms)

This paper contains 67 sections, 5 theorems, 14 equations, 4 figures, 5 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Works
  3. Multi‑Label Learning and the Long‑Tail Challenge
  4. Existing Remedies for Long‑Tail MLC
  5. Optimization‑Centric & Game‑Theoretic Views
  6. Positioning of CD‑GTMLL
  7. Methodology
  8. Long-Tail Multi-Label Formulation
  9. Notation.
  10. Head-tail split.
  11. N-player label decomposition.
  12. Global fusion.
  13. Game-Theoretic Formulation
  14. Cooperative objective.
  15. Curiosity-Driven Exploration for Tail Labels
  16. ...and 52 more sections

Key Result

Theorem 1

Under Assumption ass:continuity, there exists a global maximizer $(\theta_{1}^{*},...,\theta_{N}^{*})\in\prod_{i}\Theta_{i}$ of eq:payoff. Every such maximizer is a pure-strategy Nash equilibrium, and—because $\mathcal{M}$ is tail-responsive—no equilibrium can systematically ignore all tail labels.

Figures (4)

  • Figure 1: Precision@k (mean ± std, %) results on three benchmark datasets.
  • Figure 2: Rare-F1 (mean ± std, %) results on artificially down-sampled "-R" variants. @X% indicates the fraction of positives removed.
  • Figure 3: Mean performance rank (mean ± std) of each player on different label sets for Pascal VOC and MS-COCO. A lower rank indicates better performance.
  • Figure 4: Comparison of computational overhead and performance on the MS-COCO dataset.

Theorems & Definitions (6)

  • Theorem 1: Existence of a tail-aware equilibrium
  • Lemma 1: Bounded global payoff
  • Proposition 1: Curiosity prioritizes tail labels
  • Theorem 2
  • definition 1: Rare-F1
  • Proposition 2