Goto

Collaborating Authors

 memory footprint


We provide a simple pseudo-2

Neural Information Processing Systems

We thank all the reviewers for their constructive comments. We will provide details in the final draft. MCUNet shows consistent improvement across different devices (F746, H743) and tasks (classification, detection). R1: Whether the overall network topology brings major improvement. R2: Why the auto-tuning in TVM fails to work on MCUs.





main aim of our work is to develop reversible graph neural network models, called Graph Normalizing Flows (GNFs)

Neural Information Processing Systems

We thank the reviewers for their detailed comments. We are glad to see a generally positive assessment of our work. We will report larger-scale results in the final draft. Below, we address specific reviewer comments. When using 12G GPU machines, this difference is significant.