Concept graph
Generative AI and LLM glossary
Clear AI definitions with practical context, examples, and links to related models, tools, and concepts.
Machine Learning
Adversarial Training
Adversarial training improves model robustness by training on deliberately perturbed inputs that are designed to expose failures under a defined threat model.
Training Techniques
Backpropagation
An algorithm used for training neural networks by calculating gradients of the loss function.
Deep Learning
Batch Normalization
A technique to improve training speed and stability in deep neural networks by normalizing the inputs of each layer.
Product
Chain of thought
Chain of thought is prompting or training a model to write intermediate reasoning steps before a final answer.
Data Processing
Data Annotation
The process of labeling data for training machine learning models.
Data Processing
Data augmentation
Data augmentation creates extra training examples by transforming existing data, such as crops, paraphrases, or audio noise.
Data Management
Dataset
A structured collection of data used for analysis and training machine learning models.
Data Management
Dataset Splitting
The process of dividing a dataset into training, validation, and test sets.
Machine Learning
distributed-learning
A machine learning paradigm where the training data is distributed across multiple devices or nodes.
Regularization
Dropout
A regularization technique used to prevent overfitting in neural networks by randomly deactivating a fraction of neurons during training.
Data Processing
feature-extraction
The process of transforming raw data into a set of usable characteristics for model training.
Data Processing
feature-selection
The process of selecting a subset of relevant features for model training.
Training
Fine-tuning
Fine-tuning continues training a pretrained model on a smaller, task-specific dataset so it follows that task more reliably.
Neural Networks
Generative Adversarial Network (GAN)
A generative adversarial network (GAN) trains a generator and a discriminator in opposition so the generator learns to produce samples that resemble a training distribution.
Machine Learning
Hyperparameter tuning
Hyperparameter tuning searches training or decoding settings that are not learned as weights, such as learning rate or temperature.
Machine Learning
incremental-learning
A machine learning approach that updates models continuously with new data without retraining from scratch.
Serving
Inference
Inference is running a trained model on new input to produce an output. Training updates the model's weights. Inference does not.
Machine Learning
Model Training
The process of teaching a machine learning model to make predictions based on data.
modeling
model-trained
The process of training a machine learning model using data.
data-quality
noisy-labels
Labels in a dataset that are inaccurate or wrong, often leading to misguidance in model training.
machine learning
one-shot-learning
A machine learning approach that enables a model to learn information about object categories from a single training example.
machine learning
out-of-distribution-generalization
The ability of a model to perform well on unseen data that is different from the training set.
Modeling
Overfitting
A modeling error that occurs when a model learns noise and details from the training data.
Machine Learning
semi-supervised-learning
A learning approach that combines labeled and unlabeled data for training models.