As machine learning models in critical fields increasingly grapple with multimodal data, they face the dual challenges of handling a wide array of modalities, often incomplete due to missing elements, and the temporal irregularity and sparsity of collected samples. Successfully leveraging this complex data, while overcoming the scarcity of high-quality training samples, is key to improving these models' predictive performance. We introduce ``FuseMoE'', a mixture-of-experts framework incorporated with an innovative gating function. Designed to integrate a diverse number of modalities, FuseMoE is effective in managing scenarios with missing modalities and irregularly sampled data trajectories. Theoretically, our unique gating function contributes to enhanced convergence rates, leading to better performance in multiple downstream tasks. The practical utility of FuseMoE in the real world is validated by a diverse set of challenging prediction tasks.

A long, long time ago, in a laboratory far, far away, EDA researchers and developers used paper tape instead of Linux, rubylith instead of GDS II, yellow wires instead of ten levels of metal. Sitting around a potbellied stove, in their rocking chairs, practitioners of that era (and this) will offer insight into why some great ideas were immediately put into practice while others stayed on the drawing board or in the ivory tower. They will share remembrances of things past, of simpler days when foundries made steel, when options meant CMOS or bipolar, when real parts were measured instead of benchmarks touted. Their stories of what it was like, what has changed, and whether the "good old days" were then or now will be followed by questions and, possibly, answers.