ResNeXt

Introduced by Xie et al. in Aggregated Residual Transformations for Deep Neural Networks

A ResNeXt repeats a building block that aggregates a set of transformations with the same topology. Compared to a ResNet, it exposes a new dimension, cardinality (the size of the set of transformations) $C$, as an essential factor in addition to the dimensions of depth and width.

Formally, a set of aggregated transformations can be represented as: $\mathcal{F}(x)=\sum_{i=1}^{C}\mathcal{T}_i(x)$, where $\mathcal{T}_i(x)$ can be an arbitrary function. Analogous to a simple neuron, $\mathcal{T}_i$ should project $x$ into an (optionally low-dimensional) embedding and then transform it.

Source: Aggregated Residual Transformations for Deep Neural Networks

Read Paper See Code

Papers

Paper	Code	Results	Date	Stars

Tasks

Task	Papers	Share
Object Detection	38	12.42%
Image Classification	28	9.15%
Semantic Segmentation	23	7.52%
General Classification	19	6.21%
Instance Segmentation	15	4.90%
Object	13	4.25%
Classification	8	2.61%
Action Recognition	7	2.29%
Panoptic Segmentation	6	1.96%

Usage Over Time

This feature is experimental; we are continuously improving our matching algorithm.

Components

Component	Type	Add Remove
1x1 Convolution	Convolutions
Batch Normalization	Normalization
Convolution	Convolutions
Global Average Pooling	Pooling Operations
Grouped Convolution	Convolutions
Kaiming Initialization	Initialization
ReLU	Activation Functions
Residual Connection	Skip Connections
ResNeXt Block	Skip Connection Blocks

Categories

Add Remove

Convolutional Neural Networks