Untitled Flashcards Set

Training Set

The collection of data used to teach a machine learning model, including multiple examples for learning patterns.

Example (Instance)

One row from the training set, consisting of features (inputs) and a label (output).

Label (Target)

The value that a model is trying to predict; also known as the dependent variable.

Features (Inputs)

Input variables used to make predictions; also called independent variables.

Linear Regression

A simple machine learning model that assumes a straight-line relationship between features and labels.

Weights

Numbers that determine how important each feature is for the model's prediction.

Bias

A constant that adjusts the predictions made by the model, ensuring a non-zero output even when all features are zero.

Loss Function

A method for measuring how inaccurate a model's predictions are, commonly using Mean Squared Error.

Gradient Descent

An optimization algorithm used by models to learn by adjusting weights and bias step by step.

Stochastic Gradient Descent (SGD)

A variant of gradient descent that updates weights using one example at a time.

Minibatch Gradient Descent

A method that updates weights using a small batch of examples, balancing speed and stability.

Parameters

Values that the model learns automatically during training, including weights and bias.

Hyperparameters

Settings that are manually chosen before training; these control how the model learns but are not learned automatically.

Validation Set

A separate dataset used to evaluate the model's performance on new data, preventing overfitting.

Vectorization

Using matrices and vectors for computations to speed up calculations in machine learning.

Biological Neurons

Real neurons in the brain, which inspired the design of artificial neural networks.

Artificial Neurons

Components of artificial neural networks that mimic biological neurons to process input data.

Soma (Cell Body)

The part of a biological neuron that processes signals, comparable to a neuron (node) in an artificial neural network.

Dendrites

Extensions of a biological neuron that receive signals from other neurons, analogous to the input layer that receives input features in an artificial neural network.

Axon

The long projection from a biological neuron that sends output to other neurons, similar to the output layer producing predictions in an artificial neural network.

Synapse

The junction between biological neurons for signal transmission, akin to weights and bias in an artificial neural network which determine the importance of inputs.