Artificial Intelligence, Machine Learning and Deep Learning

Source: Deep Learning on Medium

Artificial Intelligence, Machine Learning and Deep Learning

Deep Learning vs Machine Learning vs Artificial Intelligence

“Artificial intelligence is the future. Artificial intelligence is science fiction. Artificial intelligence is already part of our everyday lives. All those statements are true, it just depends on what flavor of AI you are referring to.”

  1. Artificial Intelligence

Nowadays it is difficult to speak about machine learning without citing artificial intelligence, and it is also difficult to speak about deep learning without talk about machine learning. Before continuing in this reading it is important to make some clarifications to avoid further confusion [1].

The dream of AI pioneers of ’56 was a complex machine able to perform “human actions”. The idea of a human intelligence that have all our senses, all our reason, and think like a human is a concept that we can express as “General AI”.

Is it possible to see General AI in so many films (Star Wars, Terminator, Space Odyssey, etc.), but is (for now) too earlier to speak about such complex system.

What is possible to find currently are “Narrow AI” technologies developed for specific tasks, common examples are image classifier or face recognition.

These examples provide some facets of human intelligence, but how? Where does that intelligence come from? This takes us to the next step of the reading, Machine Learning.

2. Machine Learning

Machine learning is the practice of using algorithms to parse data, learn from it, and then make a determination or prediction about something in the world.

The goal of this process is training a software (instead of write hand-coding routines) with a specific set of instructions, amounts of data and algorithms, which give to it the ability to accomplish particular and very specific tasks.

Machine learning is inspired by the neural networks of our brains, but instead of neurons connected between them, the artificial neural network uses the concept of “layer”.
Each layer consists of one or more node that can be connected to the node of the next layer. The lines between node indicate the information flow from one node to the next one [2]. Each node assigns a weighting to its input. The final output is then determined by the total of those weightings.

Day by day it appears that one of the best application of machine learning was in the computer vision field, the deep learning.

3. Deep Learning

For deep learning is meant a vast set of algorithms that relying on the concept of neural networks, expand up to contain a large number of nodes that are scattered across depths, hence the deep learning name [3].

Today, image recognition provided by neural networks, trained via deep learning is, in some scenarios, better than humans’ recognition. For example: voice and image recognition provided by Google, use deep learning, what to buy or watch on Amazon and/or Netflix is generated by artificial neural networks trained on your taste but; how does it work? [4]

[Extract from the paper “Deep Learning and Self-Driving Cars” Elia Rigo —University of Trento]

References

1. Blogs NVIDIA, Article by Michael Copeland, https://blogs.NVIDIA.com/blog/2016/07/29/whats-difference-artificial-intelligence-machine-learning-deep-learning-ai/, last accessed 2017/10/2

2. Author, Steven W. Smith Title, Digital Signal Processing, Chapter 26 (2011)

3. MOTHERBOARD, Article by Riccardo Coluccini, https://motherboard.vice.com/it/article/pg5vby/deep-learning-futuro-intelligenza-artificiale, last accessed 2017/10/26

4. Forbes, Article by Bernard Marr, https://www.forbes.com/sites/bernardmarr/2016/12/08/what-is-the-difference-between-deep-learning-machine-learning-and-ai/#21ee325d26cf, last accessed 2017/10/28