What is the purpose of Machine Learning?

To automate tasks that typically require human computation and thinking.

What are three components needed for machine learning?

Input data, expected output, measure of if algorithm is doing well.

What is supervised learning?

Supervised Learning is when a machine learns how to map input to correct output from labelled examples

What is the difference between classification and regression?

Classification is when a machine classifies something with a categorical output, ex. a cat. Regression is when the machine classifies something with a numerical or probabilistic output, ex. price of car

What is unsupervised learning?

When there are no labelled examples, but the machine still needs to find interesting patterns in the data.

What is reinforcement learning?

When the learning system observes the environment, performs actions, and either gets a negative or positive reward signal, and learns to maximize positive reward signals.

What is causality in machine learning?

It is when a machine understands that changing a variable causes another variable to change, rather than just understanding the variables are correlated

What signs show that more data is needed for a model

If more data is needed for a model, it is likely underfitting, so it will have high variance and high testing error, but low training error, and the models predictions are unstable, with small changes in data resulting in large prediction changes

How do you know if a model will generalize well to new data

it does not overfit, has low testing error, training and validation errors are similar,

Help

Options

focusNode

Didn't know it?
click below

Knew it?
click below

Don't Know

Remaining cards (0)

Know

retry

shuffle

restart

0:00

Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.

Normal Size Small Size show me how

DS3000

Concepts I Should Recall For DS3000 at Western

Question	Answer
What is the purpose of Machine Learning?	To automate tasks that typically require human computation and thinking.
What are three components needed for machine learning?	Input data, expected output, measure of if algorithm is doing well.
What is supervised learning?	Supervised Learning is when a machine learns how to map input to correct output from labelled examples
What is the difference between classification and regression?	Classification is when a machine classifies something with a categorical output, ex. a cat. Regression is when the machine classifies something with a numerical or probabilistic output, ex. price of car
What is unsupervised learning?	When there are no labelled examples, but the machine still needs to find interesting patterns in the data.
What is reinforcement learning?	When the learning system observes the environment, performs actions, and either gets a negative or positive reward signal, and learns to maximize positive reward signals.
What is causality in machine learning?	It is when a machine understands that changing a variable causes another variable to change, rather than just understanding the variables are correlated
What signs show that more data is needed for a model	If more data is needed for a model, it is likely underfitting, so it will have high variance and high testing error, but low training error, and the models predictions are unstable, with small changes in data resulting in large prediction changes
How do you know if a model will generalize well to new data	it does not overfit, has low testing error, training and validation errors are similar,

Created by: EngineeringGirl

"Know" box contains:
Time elapsed:
Retries: