Neural Network

Artificial Intelligence

Artificial Intelligence (AI) is the field of computer science concerned with building systems that perform tasks normally requiring human intelligence, such as…

Generative AI

Generative AI refers to models that create new content — text, images, audio, video, or code — rather than only classifying or predicting labels for existing d…

Machine Learning

Machine Learning (ML) is a subfield of AI in which systems improve their performance on a task by learning patterns from data, rather than following explicitly…

Deep Learning

Deep Learning is a subset of machine learning that uses artificial neural networks with many layers ('deep' architectures) to automatically learn hierarchical…

Neural Network

A neural network is a computational model composed of interconnected layers of simple processing units called neurons, loosely inspired by biological brains. E…

Embeddings

Embeddings are dense numerical vector representations of data — such as words, sentences, images, or documents — learned so that semantically similar items are…

LLM

A Large Language Model (LLM) is a neural network, typically based on the transformer architecture, trained on massive amounts of text to predict and generate h…

Inference

Inference is the process of running a trained machine learning model on new input data to produce a prediction, classification, or generated output. It is dist…

LoRA

LoRA (Low-Rank Adaptation) is a parameter-efficient fine-tuning technique that adapts a large pretrained model by freezing its original weights and injecting s…

Pika

Pika is an AI video generation tool that creates and edits short video clips from text prompts, images, or existing video, using generative AI models.

Catastrophic Forgetting

Catastrophic forgetting is the tendency of a neural network to abruptly lose previously learned knowledge when it is trained on new data or tasks, because the…

Neural Architecture Search

Neural Architecture Search (NAS) is an automated technique for designing effective neural network architectures by systematically searching a space of possible…

Variational Autoencoder

A Variational Autoencoder (VAE) is a generative neural network that learns a probabilistic latent representation of data, enabling it to both reconstruct input…

Generative Adversarial Network

A Generative Adversarial Network (GAN) is a generative modeling framework in which two neural networks — a generator and a discriminator — are trained in compe…

Softmax Function

The softmax function converts a vector of raw real-valued scores (logits) into a probability distribution, exponentiating each score and dividing by the sum of…

Gradient Clipping

Gradient clipping is a training technique that caps the magnitude of gradients during backpropagation, preventing the exploding-gradient problem that can desta…