ML Algorithms
This algorithm is often used for binary classification tasks and works by finding the best line or hyperplane that divides data into two classes.
What is a Support Vector Machine (SVM)?
The method used for updating the weights in a neural network during training.
What is backpropagation?
In 1997, this IBM supercomputer defeated world chess champion Garry Kasparov in a highly publicized match, marking a significant moment in the history of artificial intelligence.
This type of learning occurs when an AI system learns from data that is neither labeled nor categorized, instead finding patterns and structures in the data on its own.
What is unsupervised learning?
This ensemble method builds multiple decision trees and merges them together to get more accurate and stable predictions.
What is Random Forest?
The number of output units in a neural network designed for a regression task
What is one?
In what year was the backpropagation method of training neural networks published?
What is 1986?
The name for the situation where information from the test set influences the process of model building.
What is data leakage?
This activation function has the equation 1 / (1+e^(-z))
What is the sigmoid?
The problem that arises when the gradients of some activation functions become very small to the point of being negligible, resulting in the weights failing to update.
What is the vanishing gradient problem?
The birth of the field of AI research is commonly attributed to a conference held at this university in 1956.
What is Dartmouth?
n neural networks, this term refers to a function applied to outputs at each layer that helps introduce nonlinearity in the model, allowing it to learn more complex patterns.
What is an activation function?
The name of the loss function used in logistic regression
What is the Negative Log Likelihood?
The term we optionally add to the update rule when training a neural network which is a factor of the previous weight change in order to smooth training and avoid local minima
What is momentum?
Who created the ImageNet competition which was a key driver of the current deep learning boom?
Who is Fei-Fei Li?
This graphical plot is crucial for evaluating the diagnostic ability of binary classifiers, depicting the trade-off between true positive rate and false positive rate at various threshold settings.
What is a Receiver Operating Characteristic (ROC) curve?
A form of unsupervised learning which clusters data points based on areas of density
What is DBSCAN?
The NN training strategy of a small initial learning rate increasing to a max value and then decaying back down, with momentum cycling inversely to learning rate.
What is the one-cycle policy?
Who is credited with popularizing the term Machine Learning in 1959?
Who is Arthur Samuel?
The classification error metric which answers the question "How many of all positives did the model correctly classify as positives?"
What is recall (or sensitivity)?