Your peers in the course Einführung Data Science at the Universität Stuttgart create and share summaries, flashcards, study plans and other learning materials with the intelligent StudySmarter learning app.

Get started now!

Einführung Data Science

What is a qualitative variable?

Also called categorical variables, take on **values that are names or labels** e.g. the gender

Einführung Data Science

What are discrete variables?

Can only take on a **finite number of values**, eg. The number of cars owned.

Einführung Data Science

What is an Continuous variable?

Can take on **any value in a certain range, **eg. The body height

Einführung Data Science

What is the major goal of supervised learning?

To find parameter values that will allow a model** to perform well** on historical data and to **make predictions on unknown data** that have benn part **of the training dataset**.

Einführung Data Science

What is meant by **Classification?**

- the process of
**modeling and predicting a discrete class label**that is often not measurable but observable - two classes to predict, usually 1 or 0 values, then it is called
**binary classification** - when there are more
**than two class labels**to predict it is called**multi-classification** - examples eg predicting employee churn, email spam, financial fraud, student letter grades

Einführung Data Science

What is the **problem **with hyperparameters?

There is no **magic number that work everywhere**

The best hyperparameter setting for a specific learning algorithm specifically depends on each learning task and on each dataset!

The best hyperparameter setting for a specific learning algorithm specifically depends on each learning task and on each dataset!

Einführung Data Science

What are approaches inherited from the **hyperparameter optimization**?

grid search and random search

Einführung Data Science

What fundamental problem lies ahead when searching for the **right **learning rate?

- high values for the learning rate often lead to
**overshooting or oscillation**around the minimum - very low values often
**slowing down**the process of reaching the minimum and/or tapping the algorithm in an**undesirable local minimum**

Einführung Data Science

What does **iterative optimization algorithm **mean?

to pass over the training set for a multiple of times in a way that can also be controlled via a series of additional hyperparameters in order to get the best result

Einführung Data Science

To what refers Iterations?

The **number of batches need to complete one epoch**

For a training set with 1000 instances and a batch size of 250 it would take**4 iterations **to complete **1 epoch**

For a training set with 1000 instances and a batch size of 250 it would take

Einführung Data Science

What are the main problems of the Batch Gradient Descent?

- the batch gradient descent passes over the
**entire training set**before taking**only a single small step**in the direction of steepest descent- therefore it is a
**very costly operation**especially if the training set is**large**

- therefore it is a
- another problem with
**gradient descent algorithms**in general and with the batch gradient descent**in particular**is that they are**susceptible to local minima**

Einführung Data Science

What ensures strict convexity?

That the only **local optimum of a function is global **at the same time

For your degree program Einführung Data Science at the Universität Stuttgart there are already many courses on StudySmarter, waiting for you to join them. Get access to flashcards, summaries, and much more.

Back to Universität Stuttgart overview pageStudySmarter is an intelligent learning tool for students. With StudySmarter you can easily and efficiently create flashcards, summaries, mind maps, study plans and more. Create your own flashcards e.g. for Einführung Data Science at the Universität Stuttgart or access thousands of learning materials created by your fellow students. Whether at your own university or at other universities. Hundreds of thousands of students use StudySmarter to efficiently prepare for their exams. Available on the Web, Android & iOS. It’s completely free.

Best EdTech Startup in Europe