动态调整训练数据，让模型自我调节学习节奏。
Adaptive Data Dropout: Towards Self-Regulated Learning in Deep Neural Networks
- 根据训练效果实时调整数据使用量，不靠固定规则。
- 减少有效训练步数，准确率仍优于静态方法。
- 适合追求高效训练的深度学习研究者。
深度神经网络通常在各轮次中均匀采样大规模数据集，但研究表明并非所有样本在学习过程中贡献均等。已有工作表明逐步减少训练数据量可提升效率与泛化能力，但现有方法依赖固定调度，无法在训练中动态调整。本文提出自适应数据丢弃（Adaptive Data Dropout），一种基于性能反馈动态调整训练数据子集的简单框架。受自我调节学习启发，该方法将数据选择视为自适应过程，根据训练准确率的变化增减数据暴露。我们引入轻量级随机更新机制，在线调节丢弃策略，使模型在探索与巩固间实现平衡。在标准图像分类基准上的实验表明，该方法在减少有效训练步数的同时，保持了与静态数据丢弃策略相当的竞争力。结果凸显自适应数据选择在高效、鲁棒训练中的潜力。代码将公开。
原文摘要 · Abstract (English)
Deep neural networks are typically trained by uniformly sampling large datasets across epochs, despite evidence that not all samples contribute equally throughout learning. Recent work shows that progressively reducing the amount of training data can improve efficiency and generalization, but existing methods rely on fixed schedules that do not adapt during training. In this work, we propose Adaptive Data Dropout, a simple framework that dynamically adjusts the subset of training data based on performance feedback. Inspired by self-regulated learning, our approach treats data selection as an adaptive process, increasing or decreasing data exposure in response to changes in training accuracy. We introduce a lightweight stochastic update mechanism that modulates the dropout schedule online, allowing the model to balance exploration and consolidation over time. Experiments on standard image classification benchmarks show that our method reduces effective training steps while maintaining competitive accuracy compared to static data dropout strategies. These results highlight adaptive data selection as a promising direction for efficient and robust training. Code will be released.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。