DyCE: Dynamic Configurable Exiting for Deep Learning Compression and Scaling

arXiv:2403.01695v1 Announce Type: new
Abstract: Modern deep learning (DL) models necessitate the employment of scaling and compression techniques for effective deployment in resource-constrained environments. Most existing techniques, such as pruning and quantization are generally static. On the other hand, dynamic compression methods, such as early exits, reduce complexity by recognizing the difficulty of input samples and allocating computation as needed. Dynamic methods, despite their superior flexibility and potential for co-existing with static methods, pose significant challenges in terms of implementation …

abstract arxiv complexity compression cs.ai cs.lg deep learning deployment dynamic employment environments exits modern pruning quantization reduce scaling type