理论 · online learning论文趋势 2023–2026

主要结论

• 理论 · online learning论文从 2023 年的 164 篇增长到 2026 年的 190 篇，三年间扩大到约 1.2 倍。
• NeurIPS 是发理论 · online learning论文最多的会议，2023–2026 年共 427 篇，占四个会议总量的 56%。
• 交叉方向里涨得最快的是驾驶仿真：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 12 篇。
• 在「理论与统计」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的理论 · learning theory是 5.6 倍，理论 · online learning是 1.2 倍。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• NeurIPS 2026A Single-Sample Polylogarithmic Regret Bound for Nonstationary Online Linear Programming
• ICML 2026A Perturbation Approach to Unconstrained Linear Bandits
• ICLR 2026Rethinking Data Curation in LLM Training: Online Reweighting Offers Better Generalization than Offline Methods
• CVPR 2026An Optimal Transport-driven Approach for Cultivating Latent Space in Online
• NeurIPS 2025A Learning-Augmented Approach to Online Allocation Problems
• ICML 2025A Classification View on Meta Learning Bandits
• ICLR 2025Efficient Online Pruning and Abstraction for Imperfect Information Extensive-Form Games
• CVPR 2025InteractionMap: Improving Online Vectorized HDMap Construction with Interaction
• NeurIPS 2024A Best-of-both-worlds Algorithm for Bandits with Delayed Feedback with Robustness to Excessive Delays
• ICML 2024On Online Experimentation without Device Identifiers
• ICLR 2024Improved Regret Bounds for Non-Convex Online-Within-Online Meta Learning
• CVPR 2024Dual-Enhanced Coreset Selection with Class-wise Collaboration for Online Blurry Class Incremental Learning
• NeurIPS 2023$\varepsilon$-fractional core stability in Hedonic Games.
• ICML 2023Applied Online Algorithms with Heterogeneous Predictors
• ICLR 2023Hebbian Deep Learning Without Feedback
• CVPR 2023VisFusion: Visibility-Aware Online 3D Scene Reconstruction From Videos
• NeurIPS 2026A Theory of Adversary-Directed Online Learning
• ICML 2026A Two-Layer Framework for Joint Online Configuration Selection and Admission Control
• ICLR 2026A Statistical Learning Perspective on Semi-dual Adversarial Neural Optimal Transport Solvers
• CVPR 2026Dynamic Momentum Recalibration in Online Gradient Learning

全部理论 · online learning论文

共 763 篇，第 2 / 16 页

51. ICML 2026Bilinear Bandits with Partially Observable Features
52. ICML 2026CLASP: Online learning algorithms for Convex Losses And Squared Penalties
53. ICML 2026Clipped Q-Learning: Your Value Clipping Is Secretly A Robust Operator
54. ICML 2026Contextual Slate GLM Bandits with Limited Adaptivity
55. ICML 2026Correcting Split Selection in Online Decision Trees via Anytime-Valid Inference
56. ICML 2026CUPID in the Model Zoo: Online Matchmaking for Selecting Your Dream LLM
57. ICML 2026Data- and Variance-dependent Regret Bounds for Online Tabular MDPs
58. ICML 2026Data-Source Adaptive Online Learning under Heteroscedastic Noise
59. ICML 2026Decentralized Bandits without Global Clock for Dynamic Matching Market
60. ICML 2026Decentralized Online Convex Optimization with Efficient Communication: Improved Algorithm and Lower Bounds
61. ICML 2026Doubly Outlier-Robust Online Infinite Hidden Markov Model
62. ICML 2026Efficient Online Influence Maximization under the Independent Cascade Model with Node-Level Feedback
63. ICML 2026Efficiently Learning Drifting Halfspaces with Massart Noise
64. ICML 2026Exact Unlearning in Reinforcement Learning
65. ICML 2026Fast Non-Episodic Finite-Horizon RL with K-Step Lookahead Thresholding
66. ICML 2026Finite and Corruption-Robust Regret Bounds in Online Inverse Linear Optimization under M-Convex Action Sets
67. ICML 2026Generalized Linear Bandits with Memory
68. ICML 2026Hybrid Reinforcement Learning in Adversarial Markov Decision Processes
69. ICML 2026Improved Dimension Dependence for Bandit Convex Optimization with Gradient Variation
70. ICML 2026Inference of Online Newton Methods with Nesterov's Accelerated Sketching
71. ICML 2026Keep Everyone Happy: Online Fair Division of Numerous Items with Few Copies
72. ICML 2026Knowing Who, Not How Much: Learning-Augmented Mechanisms for Consumer Utility Maximization
73. ICML 2026Learning in Bayesian Stackelberg Games With Unknown Follower's Types
74. ICML 2026Learning in Structured Stackelberg Games
75. ICML 2026Learning-Augmented Online Covering Problems
76. ICML 2026Learning-Augmented Online Minimization with Dual Predictions
77. ICML 2026Learning-augmented Rent-or-Buy with a Sample
78. ICML 2026Linear Bandits beyond Inner Product Spaces, the case of Bandit Optimal Transport
79. ICML 2026Logarithmic Switching Regret for Online Convex Optimization
80. ICML 2026MindZero: Learning Online Mental Reasoning With Zero Annotations
81. ICML 2026Minimax Optimal Strategy for Delayed Observations in Online Reinforcement Learning
82. ICML 2026Minimax-Optimal Policy Regret in Partially Observable Markov Games
83. ICML 2026Modality-Decoupled Online Recursive Editing
84. ICML 2026Near-Optimal Regret for KL-Regularized Multi-Armed Bandits
85. ICML 2026Neural Logistic Bandits
86. ICML 2026Next-Token Prediction and Regret Minimization
87. ICML 2026Non-Adversarial Imitation Learning Provably Free of Compounding Errors: The Value Flow Mechanism
88. ICML 2026Non-Stationary Online Structured Prediction with Surrogate Losses
89. ICML 2026On Contraction of Sequential and Offset Rademacher Complexities
90. ICML 2026On the Coordination of Value-Maximizing Bidders
91. ICML 2026On the Role of Computation in Reinforcement Learning
92. ICML 2026Online Compatible Reward Identification from Preference Feedback
93. ICML 2026Online Conformal Prediction via Universal Portfolio Algorithms
94. ICML 2026Online Continual Learning with Dynamic Label Hierarchies
95. ICML 2026Online Contract Design With Unknown Technology
96. ICML 2026Online Fair Division with Additional Information
97. ICML 2026Online Learning with Recency: Algorithms for Sliding-window Streaming Multi-armed Bandits
98. ICML 2026Online Packet Scheduling with Deadlines and Learning
99. ICML 2026Online Rubrics Elicitation from Pairwise Comparisons
100. ICML 2026Online Social Welfare Function-based Resource Allocation

常见问题

2023–2026 年四大会议一共有多少篇理论 · online learning论文？

共 763 篇，按年份分别是 2023 年 164 篇、2024 年 224 篇、2025 年 185 篇、2026 年 190 篇。

NeurIPS 2026 有多少篇理论 · online learning论文？

NeurIPS 2026 有 68 篇理论 · online learning论文，上一届是 89 篇。

ICLR 2026 有多少篇理论 · online learning论文？

ICLR 2026 有 27 篇理论 · online learning论文，上一届是 18 篇。

ICML 2026 有多少篇理论 · online learning论文？

ICML 2026 有 86 篇理论 · online learning论文，上一届是 74 篇。

CVPR 2026 有多少篇理论 · online learning论文？

CVPR 2026 有 9 篇理论 · online learning论文，上一届是 4 篇。

哪个会议的理论 · online learning论文最多？

NeurIPS，2023–2026 年共 427 篇。