PolyConv learns continuous distributions as the convolutional filters to share the weights across different vertices of graphs or points of point clouds.

PolyConv

PolyNet: Polynomial Neural Network for 3D Shape Recognition with PolyShape Representation

**DCGAN**, or **Deep Convolutional GAN**, is a generative adversarial network architecture. It uses a couple of guidelines, in particular:

- Replacing any pooling layers with strided convolutions (discriminator) and fractional-strided convolutions (generator).
- Using batchnorm in both the generator and the discriminator.
- Removing fully connected hidden layers for deeper architectures.
- Using [ReLU](https://paperswithcode.com/method/relu) activation in generator for all layers except for the output, which uses tanh.
- Using LeakyReLU activation in the discriminator for all layer.

DCGAN

Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks

**BiSeNet V2** is a two-pathway architecture for real-time semantic segmentation. One pathway is designed to capture the spatial details with wide channels and shallow layers, called Detail Branch. In contrast, the other pathway is introduced to extract the categorical semantics with narrow channels and deep layers, called Semantic Branch. The Semantic Branch simply requires a large receptive field to capture semantic context, while the detail information can be supplied by the Detail Branch. Therefore, the Semantic Branch can be made very lightweight with fewer channels and a fast-downsampling strategy. Both types of feature representation are merged to construct a stronger and more comprehensive feature representation.

BiSeNet V2

BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation

**SPP-Net** is a convolutional neural architecture that employs [spatial pyramid pooling](https://paperswithcode.com/method/spatial-pyramid-pooling) to remove the fixed-size constraint of the network. Specifically, we add an SPP layer on top of the last convolutional layer. The SPP layer pools the features and generates fixed-length outputs, which are then fed into the fully-connected layers (or other classifiers). In other words, we perform some information aggregation at a deeper stage of the network hierarchy (between convolutional layers and fully-connected layers) to avoid the need for cropping or warping at the beginning.

SPP-Net

Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition

**Wasserstein GAN**, or **WGAN**, is a type of generative adversarial network that minimizes an approximation of the Earth-Mover's distance (EM) rather than the Jensen-Shannon divergence as in the original [GAN](https://paperswithcode.com/method/gan) formulation. It leads to more stable training than original GANs with less evidence of mode collapse, as well as meaningful curves that can be used for debugging and searching hyperparameters.

WGAN

Wasserstein GAN

**ManifoldPlus** is a method for robust and scalable conversion of triangle soups to watertight manifolds. It extracts exterior faces between occupied voxels and empty voxels, and uses a projection based optimization method to accurately recover a watertight manifold that resembles the reference mesh. It does not rely on face normals of the input triangle soups and can accurately recover zero-volume structures. For scalability, it employs an adaptive Gauss-Seidel method for shape optimization, in which each step is an easy-to-solve convex problem.

ManifoldPlus

ManifoldPlus: A Robust and Scalable Watertight Manifold Surface Generation Method for Triangle Soups

**Contextual Residual Aggregation**, or **CRA**, is a module for image inpainting. It can produce high-frequency residuals for missing contents by weighted aggregating residuals from contextual patches, thus only requiring a low-resolution prediction from the network. Specifically, it involves a neural network to predict a low-resolution inpainted result and up-sample it to yield a large blurry image. Then we produce the high-frequency residuals for in-hole patches by aggregating weighted high-frequency residuals from contextual patches. Finally, we add the aggregated residuals to the large blurry image to obtain a sharp result.

Contextual Residual Aggregation

Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting

**Movement Pruning** is a simple, deterministic first-order weight pruning method that is more adaptive to pretrained model fine-tuning. Magnitude pruning can be seen as utilizing zeroth-order information (absolute value) of the running model. In contrast, movement pruning methods are where importance is derived from first-order information. Intuitively, instead of selecting weights that are far from zero, we retain connections that are moving away from zero during the training process.

Movement Pruning

Movement Pruning: Adaptive Sparsity by Fine-Tuning

The method exploits the finding that high correlation of segmentation performance among each U-Net's decoder layer -- with discriminative layer attached -- tends to have higher segmentation performance in the final segmentation map. By introducing an "Inter-layer Divergence Loss", based on Kulback-Liebler Divergence, to promotes the consistency between each discriminative output from decoder layers by minimizing the divergence.

If we assume that each decoder layer is equivalent to PDE function parameterized by weight parameter $\theta$:

$Decoder_i(x;\theta_i) \equiv PDE(x;\theta_i)$

Then our objective is trying to make each discriminative output similar to each other:

$PDE(x; \theta_d) \sim PDE(x; \theta_i);\text{ } 0 \leq i < d$

Hence the objective is to $\text{minimize} \sum_{i=0}^{d} D_{KL}(\hat{y} || Decoder_i)$.

SSDS

OCTAve: 2D en face Optical Coherence Tomography Angiography Vessel Segmentation in Weakly-Supervised Learning with Locality Augmentation

**Feature Pyramid Grids**, or **FPG**, is a deep multi-pathway feature pyramid, that represents the feature scale-space as a regular grid of parallel bottom-up pathways which are fused by multi-directional lateral connections. It connects the backbone features, $C$, of a ConvNet with a regular structure of $p$ parallel top-down pyramid pathways which are fused by multi-directional lateral connections, AcrossSame, AcrossUp, AcrossDown, and AcrossSkip. AcrossSkip are direct connections while all other types use [convolutional](https://paperswithcode.com/method/convolution) and [ReLU](https://paperswithcode.com/method/relu) layers.

On a high-level, FPG is a deep generalization of [FPN](https://paperswithcode.com/method/fpn) from one to $p$ pathways under a dense lateral connectivity structure.

Feature Pyramid Grids

**ProxylessNet-GPU** is a convolutional neural network architecture learnt with the [ProxylessNAS](https://paperswithcode.com/method/proxylessnas) [neural architecture search](https://paperswithcode.com/method/neural-architecture-search) algorithm that is optimized for GPU devices. It uses inverted residual blocks (MBConvs) from [MobileNetV2](https://paperswithcode.com/method/mobilenetv2) as its basic building block.

ProxylessNet-GPU

ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware

The **Enhanced Fusion Framework** proposes three different ideas to improve the existing MI-based BCI frameworks.

Image source: [Fumanal-Idocin et al.](https://arxiv.org/pdf/2101.06968v1.pdf)

Enhanced Fusion Framework

Motor-Imagery-Based Brain Computer Interface using Signal Derivation and Aggregation Functions

**PAFNet** is an anchor-free detector for object detection that removes pre-defined anchors and regresses the locations directly, which can achieve higher efficiency. The overall network is composed of a backbone, an up-sampling module, an AGS module, a localization branch and a regression branch. Specifically,  ResNet50-vd is chosen as the backbone for server side, and [MobileNetV3](https://paperswithcode.com/method/mobilenetv3) for mobile side. Besides, for mobile side, we replace traditional [convolution](https://paperswithcode.com/method/convolution) layers with lite convolution operators.

PAFNet

PAFNet: An Efficient Anchor-Free Object Detector Guidance

**context2vec** is an unsupervised model for learning generic context embedding of wide sentential contexts, using a bidirectional [LSTM](https://paperswithcode.com/method/lstm). A large plain text corpora is trained on to learn a neural model that embeds entire sentential contexts and target words in the same low-dimensional space, which
is optimized to reflect inter-dependencies between targets and their entire sentential context as a whole. 

In contrast to word2vec that use context modeling mostly internally and considers the target word embeddings as their main output, the focus of context2vec is the context representation. context2vec achieves its objective by assigning similar embeddings to sentential contexts and their associated target words.

context2vec

context2vec: Learning Generic Context Embedding with Bidirectional LSTM

UNet++ is an architecture for semantic segmentation based on the [U-Net](https://paperswithcode.com/method/u-net). Through the use of densely connected nested decoder sub-networks, it enhances extracted feature processing and was reported by its authors to outperform the U-Net in [Electron Microscopy (EM)](https://imagej.net/events/isbi-2012-segmentation-challenge), [Cell](https://acsjournals.onlinelibrary.wiley.com/doi/full/10.1002/cncy.21576), [Nuclei](https://www.kaggle.com/c/data-science-bowl-2018), [Brain Tumor](https://paperswithcode.com/dataset/brats-2013-1), [Liver](https://paperswithcode.com/dataset/lits17) and [Lung Nodule](https://paperswithcode.com/dataset/lidc-idri) medical image segmentation tasks.

UNet++

UNet++: A Nested U-Net Architecture for Medical Image Segmentation

**SGDW** is a stochastic optimization technique that decouples [weight decay](https://paperswithcode.com/method/weight-decay) from the gradient update:

$$ g\_{t} =  \nabla{f\_{t}}\left(\theta\_{t-1}\right) + \lambda\theta\_{t-1}$$

$$ m\_{t} =  \beta\_{1}m\_{t-1} + \eta\_{t}\alpha{g}\_{t}$$

$$ \theta\_{t} = \theta\_{t-1} - m\_{t} - \eta\_{t}\lambda\theta\_{t-1}$$

SGDW

Decoupled Weight Decay Regularization

A resnet-like architecture with deconvolution feature normalization (Ye et al. 2020, ICLR) layers in the first few layers for sparse low-level feature identification, and batch normalization layers in the later layers.

hdxresnet

Predicting galaxy spectra from images with hybrid convolutional neural networks

A **Connectionist Temporal Classification Loss**, or **CTC Loss**, is designed for tasks where we need alignment between sequences, but where that alignment is difficult - e.g. aligning each character to its location in an audio file. It calculates a loss between a continuous (unsegmented) time series and a target sequence. It does this by summing over the probability of possible alignments of input to target, producing a loss value which is differentiable with respect to each input node. The alignment of input to target is assumed to be “many-to-one”, which limits the length of the target sequence such that it must be $\leq$ the input length.

Viet-Anh on Software

AI Glossary

Tech Terminology

Polynomial Convolution

Deep Convolutional GAN