Global Average Pooling
Global Average Pooling is a technique in convolutional neural networks (CNNs) used to down-sample feature maps by taking the average of each feature map channel.
Global Max Pooling
Global Max Pooling is a neural network operation that extracts the maximum value from each feature map channel, reducing dimensionality.
GloVe
GloVe is a word embedding technique that captures semantic relationships between words in a given corpus.
Glow
Glow is the soft, radiant light emitted by an object or surface, often used to describe a warm and inviting atmosphere.
GoogLeNet
GoogLeNet is a deep convolutional neural network architecture known for its inception modules, which allow it to efficiently process complex image data.
Gopher
Gopher is a protocol designed for distributing, searching, and retrieving documents over the Internet.
Grad-CAM
Grad-CAM is a visualization technique used in deep learning to highlight the important regions in an image that influence a model's prediction.
Gradient Clipping
Gradient clipping is a technique used in machine learning to prevent the gradients from becoming too large during training.
Gradient Descent
Gradient descent is an optimization algorithm used to minimize errors by iteratively adjusting model parameters in the direction that reduces loss.
Gram Matrix
A Gram matrix is a matrix of dot products of vectors in a given vector space.
Graph Attention Network
A Graph Attention Network (GAT) is a type of neural network architecture designed to operate on graph-structured data by leveraging attention mechanisms to focus on important nodes and edges.
Graph Convolutional Network
A Graph Convolutional Network (GCN) is a type of neural network designed to operate on graph-structured data.
Graph Isomorphism Network
Graph Isomorphism Network (GIN) is a type of neural network designed to tackle graph classification tasks by leveraging the concept of graph isomorphism to capture graph structures.
Graph Neural Network
A Graph Neural Network (GNN) is a type of neural network designed to perform inference on data structured as graphs.
Group Normalization
Group Normalization is a technique used in machine learning to stabilize and accelerate the training of neural networks by normalizing feature activations across groups of channels.
GShard
GShard is a system for efficiently scaling large-scale machine learning models by enabling model parallelism across thousands of devices.
Haiku
A haiku is a traditional Japanese poetic form consisting of three lines with a 5-7-5 syllable structure.
Hard Swish
Hard Swish is a variant of the Swish activation function used in neural networks, characterized by its piecewise linearity and computational efficiency.
He Initialization
Initialization Short Definition: Initialization is the process of assigning an initial value to a variable or data structure in programming.
Hindsight Experience Replay
Hindsight Experience Replay is a reinforcement learning technique that improves learning efficiency by utilizing unsuccessful experiences as valuable learning data.
Huber Loss
Huber Loss is a loss function used in robust regression that is less sensitive to outliers in data.
HuBERT
HuBERT is a self-supervised learning model for speech representation developed by Facebook AI, designed to improve automatic speech recognition systems.
Image Augmentation
Image augmentation is a technique used to enhance the diversity of training data in computer vision by applying various transformations to existing images.
Implicit Neural Representation
Implicit Neural Representation is a method of representing complex data using neural networks to encode continuous functions over space and time.
Page 5 of 12