The science and engineering of making intelligent machines
What is AI?
A model similar to linear regression that takes numerical inputs and produces numerical outputs.
Neural Network
Used in computer vision
What is Convolutional Neural Network?
A field that combines linguistics with computer science and Artificial Intelligence
What is Natural Language Processing?
A measure to judge how good or bad a model is
What is loss?
An activation function that turns negative numbers into 0.
What is ReLu?
It has set weights and multiplies them by pixel values and then sums everything
What is Filter?
Designed to work with sequential data - data in which the order of elements in a sequence matters
What is Recurrent Neural Network?
The algorithm learns from a “labeled” set of data, and then makes predictions on a new “unlabeled” set of data
What is supervised learning?
Brings nonlinearity to models
What is Activation function?
Takes the max number from a set of values
What is Max Pooling?
Words like “a”, “the”, “is”, “are”, etc
What are Stop Words?
The data is not given a label and the machine tries to learn on its own.
What is Unsupervised Learning?
Algorithm for calculating the gradient of a loss function with respect to variables of a model
What is Backpropagation?
A very common activation function used in a cnn
What is Softmax?
Special type of recurrent neural network designed to rectify the problem of “short-term” memory
LSTM
Calculating a model’s loss on data and adjusting the weights and biases to minimize the loss
What is Training?
A hyperparameter that determines how big or small steps are taken during gradient descent
What is learning rate?
Hyperparameter for the number of pixels by which the kernal moves after each convolutional operation
What is Stride?
For each word in the English language, computer scientists have developed a numerical vector using a model
What is Word2vec?