AI Short Introduction

Neural Networks

We often forget about neural networks. Please pay attention (and make sure you also understand the mathematical formula in the article) as it explains decision-making very well.

NNAs (neural network algorithms) use the human brain’s neurons as the basis. In simple terms, neurons allow the flow of information in our brain and body. NNAs use this principle with nodes and layers. The objective is to mimic the human brain.

But we can put this much simpler. An NNA uses units, think of these as neurons, which get collected in a layer that connects with other layers. It captures the input and transmits it to a layer output. A layer is weighted, just like our brain neuron connections. Meaning some information is less important than others. For now, think of this as a way to make a machine think more like a human. In the following article, they show a great example of how the ANN makes a decision using mathematics.

https://www.ibm.com/cloud/blog/ai-vs-machine-learning-vs-deep-learning-vs-neural-networks

Deep Learning

This is based on machine learning and can be thought of as a type of machine learning. The primary difference is that machine learning uses more complex algorithms. A term that’s used is Scalable Machine Learning (SML). This must not be confused with Supervised Machine Learning.

Given DL uses raw data, it requires a much larger amount of data.

Machine Learning

There are two types of machine learning:

• Supervised
• Unsupervised

SML uses labeled data to organise and learn.

Unsupervised does not need labels and will make sense of unstructured data. Unsupervised machine learning merely needs little data. Deep Learning is intended to mimic the brain and requires massive sets of data.

We offer contracting and freelance services and training in the field of multimodal design with the objective to create experiences your users will love.