Search Results for author:
Found 4 papers, 3 papers with code
Tiny Models are the Computational Saver for Large Models
By searching and employing the most appropriate tiny model as the computational saver for a given large model, the proposed approaches work as a novel and generic method to model compression.
Ranked #163 on
Image Classification
on ImageNet
DyCE: Dynamic Configurable Exiting for Deep Learning Compression and Scaling
Moreover, most current dynamic compression designs are monolithic and tightly integrated with base models, thereby complicating the adaptation to novel base models.
Unsupervised Pre-Training Using Masked Autoencoders for ECG Analysis
Unsupervised learning methods have become increasingly important in deep learning due to their demonstrated large utilization of datasets and higher accuracy in computer vision and natural language processing tasks.
POCKET: Pruning Random Convolution Kernels for Time Series Classification from a Feature Selection Perspective
In recent years, two competitive time series classification models, namely, ROCKET and MINIROCKET, have garnered considerable attention due to their low training cost and high accuracy.
