arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2409.02410cs.LG
收藏

自适应类别涌现训练:通过渐进式目标演化增强神经网络稳定性与泛化能力

Adaptive Class Emergence Training: Enhancing Neural Network Stability and Generalization through Progressive Target Evolution

Hassan First University of Settat(塞塔特哈桑第一大学);Faculté Sciences et Techniques(科学与技术学院)

Jaouad Dabounou

AI总结:

针对静态 one-hot 目标导致优化不稳和泛化受限的问题,本文提出目标从零向量渐进演化为 one-hot 向量的 Adaptive Class Emergence Training,并在合成与真实数据集上验证其更快收敛、更准确稳健。

摘要:

人工智能,尤其是深度神经网络的最新进展,拓展了复杂任务中可实现目标的边界。用于分类问题的传统神经网络训练方法通常依赖静态目标输出,例如 one-hot 编码向量,这可能导致优化不稳定,并难以处理数据中的非线性。本文提出一种新颖的训练方法,在整个训练过程中将目标输出从零向量逐步演化为 one-hot 编码向量。这种渐进过渡使网络能够更平滑地适应分类任务不断增加的复杂性,同时保持一种平衡状态,从而降低过拟合风险并增强泛化能力。我们的方法受有限元分析中结构平衡概念的启发,并已在合成数据集和真实世界数据集上通过大量实验得到验证。结果表明,该方法实现了更快的收敛、更高的准确率和更好的泛化,尤其是在数据复杂性和噪声较高的场景中。这一渐进式训练框架为经典方法提供了一种稳健替代方案,也为更高效、更稳定的神经网络训练开辟了新视角。

英文摘要:

Recent advancements in artificial intelligence, particularly deep neural networks, have pushed the boundaries of what is achievable in complex tasks. Traditional methods for training neural networks in classification problems often rely on static target outputs, such as one-hot encoded vectors, which can lead to unstable optimization and difficulties in handling non-linearities within data. In this paper, we propose a novel training methodology that progressively evolves the target outputs from a null vector to one-hot encoded vectors throughout the training process. This gradual transition allows the network to adapt more smoothly to the increasing complexity of the classification task, maintaining an equilibrium state that reduces the risk of overfitting and enhances generalization. Our approach, inspired by concepts from structural equilibrium in finite element analysis, has been validated through extensive experiments on both synthetic and real-world datasets. The results demonstrate that our method achieves faster convergence, improved accuracy, and better generalization, especially in scenarios with high data complexity and noise. This progressive training framework offers a robust alternative to classical methods, opening new perspectives for more efficient and stable neural network training.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑