Model compression and acceleration for deep neural networks: The principles, progress, and challenges
In recent years, deep neural networks (DNNs) have received increased attention, have been
applied to different applications, and achieved dramatic accuracy improvements in many …
applied to different applications, and achieved dramatic accuracy improvements in many …
Recent advances in convolutional neural network acceleration
In recent years, convolutional neural networks (CNNs) have shown great performance in
various fields such as image classification, pattern recognition, and multi-media …
various fields such as image classification, pattern recognition, and multi-media …
Fnet: Mixing tokens with fourier transforms
We show that Transformer encoder architectures can be sped up, with limited accuracy
costs, by replacing the self-attention sublayers with simple linear transformations that" mix" …
costs, by replacing the self-attention sublayers with simple linear transformations that" mix" …
Monarch mixer: A simple sub-quadratic gemm-based architecture
Abstract Machine learning models are increasingly being scaled in both sequence length
and model dimension to reach longer contexts and better performance. However, existing …
and model dimension to reach longer contexts and better performance. However, existing …
A survey of model compression and acceleration for deep neural networks
Deep neural networks (DNNs) have recently achieved great success in many visual
recognition tasks. However, existing deep neural network models are computationally …
recognition tasks. However, existing deep neural network models are computationally …
Monarch: Expressive structured matrices for efficient and accurate training
Large neural networks excel in many domains, but they are expensive to train and fine-tune.
A popular approach to reduce their compute or memory requirements is to replace dense …
A popular approach to reduce their compute or memory requirements is to replace dense …
Hypernetworks
This work explores hypernetworks: an approach of using a one network, also known as a
hypernetwork, to generate the weights for another network. Hypernetworks provide an …
hypernetwork, to generate the weights for another network. Hypernetworks provide an …
Recent advances in convolutional neural networks
In the last few years, deep learning has led to very good performance on a variety of
problems, such as visual recognition, speech recognition and natural language processing …
problems, such as visual recognition, speech recognition and natural language processing …
Generalisation error in learning with random features and the hidden manifold model
We study generalised linear regression and classification for a synthetically generated
dataset encompassing different problems of interest, such as learning with random features …
dataset encompassing different problems of interest, such as learning with random features …
Modeling the influence of data structure on learning in neural networks: The hidden manifold model
Understanding the reasons for the success of deep neural networks trained using stochastic
gradient-based methods is a key open problem for the nascent theory of deep learning. The …
gradient-based methods is a key open problem for the nascent theory of deep learning. The …