DropIT: Dropping Intermediate Tensors for Memory-Efficient DNN Training.

Joya Chen, Kai Xu, Yuhui Wang, Yifei Cheng, Angela Yao

Browse the full ICLR paper archive.