Machine Learning Module 2 Flashcards
Coursera Courses
What the are 2 types of Supervised Learning?
Classification and Regression
What is Classification?
Classification is the process of predicting discrete class labels or categories. i.e. Will I pass or fail my biology exam?
What is Regression?
Regression is the relationship between a dependent and independent variable. Regression is the process of predicting continous values.
Difference between Classification and Regression?
Classification works with classes while regression is mapped to a continous variable
What are some commong algorithms for regression?
Random forests, gradient boosting, support vector regression(SVR) and neural networks
What is Unsupervised learning?
The model is not supervised and allowed to work independently to discover patterns and structures in the data that is not visible to the human eye. This means the model trains on the data set and draws conclusions on its own.
Types of Unsupervised learning?
Dimension reduction, Density estimation, Market basket analysis and Clustering.
What is Dimension Reduction?
Also known as feature selection, seeks to reduce redundant features to make the classification easier.
What is Density Estimation?
Straight forward concept to explore data and find a structure
What is Market Basket Analysis?
Modelling technique based on the theory that if you buy a group of items you are more likely to buy another group of items
What is Clustering?
Most popular unsupervised learning technique and is the Grouping of Data Points or Objects that are similar by the characteristics of the data.
What is Clustering used for?
To detect Structure, to Summarize and to detect anomalies within the dataset.
What is the main difference between Supervised Learning and Unsupervised Learning?
Supervised learning deals with ‘Labelled’ data and Unsupervised learning deals with ‘Un-Labelled’ data.
What is a classifier?
A machine learning algorithm used to solve the classification problem.
What is a feature?
In independent variable that is used as an input into the model
What is evaluation?
Assessing how well a model has performed.
What are 2 types of Classification Algorithms?
Lazy learner and Eager learner
Describe the Eager learner classification model
Spends a lot of time training and generalizing the model
Describe the Lazy learner classification model
Doesn’t have a training phase per say and waits to have a test data set before making predictions. So it takes longer to predict as it doesn’t generalize the model.
What is an example of a lazy learner classification algorithm?
K-Nearest Neighbours or KNN.
How does KNN work?
KNN classifies the unknown data points by finding the most common classes in the K nearest examples. Then finds the closest match to the test data.
What is an example of an eager learning classification algorithm?
Logistic regression that is used to predict the probability of a class based on a graph plot
What is another example of an eager learning classification algorithm?
Decision trees, that are a tree like algorithm that use an ‘If’ ‘Then’ rule or set of rules. So it classifies if you pass or fail based on rules.
Other classificaiton algorithms
Support vector machines, discriminate analysis, naive Bayes, and neural networks.
What is a confusion matrix?
A confusion matrix measures the performance and accuracy of a classification problem
How is the mean squared error helpful?
The mean squared error is helpful for evalutating regression models
What is the R-squared value?
The R-squared value represents the amount of variance in a dependent variable that can be explained by an independent variable
What can be used to evauluate classificaiton models?
A confusion matrix, precision and recall
What does train/test split mean?
This is the process of splitting data into ‘training’ and ‘testing’. Some data is used to train the machine learning model and the rest is used to test it.
What is Deep Learning?
Deep Learning is a subset of machine learning and has 3 layers, Input, Hidden and Output. Deep learning is loosely based on the human brain.
What is Artificial Neural Network?
It is a group of connected nodes called neurons. ANN loosely models the human brain
What is a Neuron?
A neuron is a computation unit that can be expressed as a weighted sum of inputs. Each neuron has an activation function that predicts the value of the output.
What is a perceptron?
The coinfiguration of all inputs and their weights is a perceptron. It also contains a summation function which is a total summation of the inputs and their associated weights.
What is a Single-Layer neural network?
A neural network with one neuron.
What is a ‘Multi-Layer Perceptron?
A neural network with an input layer, one processing layer(or hidden layer) and an output layer
What is a Deep Neural Network?
An artificial neural network that has two or more hidden layers or two or more processin layers.
What is Reinforcement Learning?
Reinforcement learning is a machine learning method. Reinforcement learning works like a reward system.
What are the concepts of Reinforcement Learning?
The most common concepts in reinforcement learning are the agent, environment, reward, action, state, and policy.
What are the types of Reinforcement Learning?
The two types of reinforcement learning are positive and negative reinforcement.
What industries use reinforcement learning?
Reinforcement learning is used in various industries like games, fleet management, and manufacturing.
In reinforcement learning what is a Policy?
A policy defines how the agent will act when in a specific state, it maps the states in the environment to a single action or probability for each action. A policy can be represented as a lookup table, a function, an algorithm, or by calculating the probability of performing an action given a state.
What are the 3 algorithms for reinforcement learning?
Model based, Policy based and Value based
What is the Model based algorithm for reinforcement learning?
Here, you create a virtual model to help the agent learn in specific environments
What is the Policy based algorithm for reinforcement learning?
You develop a strategy that helps to gain maximum rewards in the future through possible actions performed in each state.
What is the Value based algorithm for reinforcement learning?
The agent’s goal here is to maximize a defined value function.
What are value functions?
Value functions are estimates for actions and states resulting from interacting with the environment.
What is Generative AI
Generative Artificial Intelligence, or GenAI, is an AI technique capable of creating new and unique data, ranging from images and music to text and entire virtual worlds.
How does Generative AI work?
Generative AI models use deep learning techniques and rely on vast datasets to generate entirely new data with various applications.
