论文标题
通过竞争性联合储层计算，对动态系统的持续学习
Continual Learning of Dynamical Systems with Competitive Federated Reservoir Computing
论文作者
论文摘要
最近，机器学习在学习微分方程和数据中的动态系统方面被证明有效。但是，通常认为数据源自单个永不改变的系统。相反，当对现实世界动态过程进行建模时，数据分布通常由于基础系统动力学的变化而发生变化。对这些过程的持续学习旨在迅速适应突然的系统变化，而不会忘记以前的动态制度。这项工作提出了一种基于储层计算的持续学习方法，储层计算是一种针对复杂时空动力系统训练复发性神经网络的最先进方法。储层计算修复了经常性的网络权重 - 因此，这些重量无法忘记 - 仅将线性投影头更新到输出。我们建议同时训练多个竞争预测负责人。受神经科学的预测编码的启发，只有最预测的头部激活，横向抑制，从而保护不活动的头部免受干扰参数更新引起的忘记。我们表明，这种多头储层可最大程度地减少在几种动态系统上的干扰和灾难性遗忘，包括Van-der-Pol振荡器，混沌Lorenz吸引人和高维Lorenz-96天气模型。我们的结果表明，储层计算是一个有前途的候选框架，用于持续学习动态系统。我们在\ url {https://github.com/leonardbereska/multiheadheadheadreservoir}上提供数据生成，方法和比较的代码。
Machine learning recently proved efficient in learning differential equations and dynamical systems from data. However, the data is commonly assumed to originate from a single never-changing system. In contrast, when modeling real-world dynamical processes, the data distribution often shifts due to changes in the underlying system dynamics. Continual learning of these processes aims to rapidly adapt to abrupt system changes without forgetting previous dynamical regimes. This work proposes an approach to continual learning based on reservoir computing, a state-of-the-art method for training recurrent neural networks on complex spatiotemporal dynamical systems. Reservoir computing fixes the recurrent network weights - hence these cannot be forgotten - and only updates linear projection heads to the output. We propose to train multiple competitive prediction heads concurrently. Inspired by neuroscience's predictive coding, only the most predictive heads activate, laterally inhibiting and thus protecting the inactive heads from forgetting induced by interfering parameter updates. We show that this multi-head reservoir minimizes interference and catastrophic forgetting on several dynamical systems, including the Van-der-Pol oscillator, the chaotic Lorenz attractor, and the high-dimensional Lorenz-96 weather model. Our results suggest that reservoir computing is a promising candidate framework for the continual learning of dynamical systems. We provide our code for data generation, method, and comparisons at \url{https://github.com/leonardbereska/multiheadreservoir}.
