<ul data-eligibleForWebStory="true"><li>A study explores hierarchical meta-learning in dynamical system reconstruction (DSR) using a Mixture of Experts (MoE) approach.</li><li>While conventional MoEs faced challenges in hierarchical DSR due to slow updates and conflicted routing, a new method called MixER is introduced.</li><li>MixER, a sparse top-1 MoE layer, incorporates a custom gating update algorithm based on $K$-means and least squares for more effective training and scalability.</li><li>Experiments validate MixER's efficiency and scalability in handling systems with up to ten parametric ordinary differential equations.</li><li>However, MixER falls short compared to existing meta-learners in scenarios with abundant data, especially when each expert processes only a fraction of a dataset with closely related data points.</li><li>Analysis with synthetic and neuroscientific time series data indicates that MixER's performance is influenced by the presence of hierarchical structure in the data.</li></ul>

Towards Foundational Models for Dynamical System Reconstruction: Hierarchical Meta-Learning via Mixture of Experts

Discover more