Skip to main content
  1. Emner/

Hvad er maskinlæring

maskinlæring er en gren af kunstig intelligens (AI) og datalogi der fokuserer på brugen af data og algoritmer til at efterligne den måde som levende væsener lærer på.

Med maskinlæring kan en computer blive bedre og bedre til at løse bestemte opgaver, jo computeren arbejder med opgaven og fodres med yderligere data.

Metoder #

Supervised In supervised learning, a data set includes its desired outputs (or labels) such that a function can calculate an error for a given prediction. The supervision comes when a prediction is made and an error produced (actual vs. desired) to alter the function and learn the mapping.

Unsupervised In unsupervised learning, a data set doesn’t include a desired output; therefore, there’s no way to supervise the function. Instead, the function attempts to segment the data set into “classes” so that each class contains a portion of the data set with common features.


Reinforcement reinforcement learning, the algorithm attempts to learn actions for a given set of states that lead to a goal state. An error is provided not after each example (as is the case for supervised learning) but instead on receipt of a reinforcement signal (such as reaching the goal state). This behavior is similar to human learning, where feedback isn’t necessarily provided for all actions but when a reward is warranted.

Neurale netværk #

A neural network processes an input vector to a resulting output vector through a model inspired by neurons and their connectivity in the brain. The model consists of layers of neurons interconnected through weights that alter the importance of certain inputs over others. Each neuron includes an activation function that determines the output of the neuron (as a function of its input vector multiplied by its weight vector). The output is computed by applying the input vector to the input layer of the network, then computing the outputs of each neuron through the network (in a feed-forward fashion).


Deep learning

Er neurale netværk som er “dybe”, det vil sige de indeholder mange lag i den matematiske model som data løber ingennem i læringsprocessen.

Bruges inden for NLP.

Anvendelse #

Speech recognition: It is also known as automatic speech recognition (ASR), computer speech recognition, or speech-to-text, and it is a capability which uses natural language processing (NLP) to translate human speech into a written format. Many mobile devices incorporate speech recognition into their systems to conduct voice search—e.g. Siri—or improve accessibility for texting.