Source: Deep Learning on Medium
How to explain Machine Learning to a grandmother who doesn’t know anything about it?
Machine learning is a trending topic actually for those people that are related with the world of technology, artificial intelligence and business, but before to start to talk about it and understand its ideas, it’s important to know from where this boom is coming, talks about its origins and history, and explain the definition of some basics concepts.
Artificial intelligence (AI): There is no exactly definition for this, but AI has 4 differents approaches who are: Thinking Humanly, Acting Humanly, Thinking Rationally, Acting Rationally, in other words, It is intended that machines think and act like human beings.
Algorithms: A sequence of instructions used to solve a problem.
Data: are individual units of information and used for the purpose of analysis.
Machine learning (ML): Algorithms that generates algorithms, that parse data, learn from that data, and then apply what they’ve learned to make informed decisions.
Neuronal Network: is a type of machine learning that follows the model of the human brain, creating an artificial neural network that through an algorithm allows the computer to learn by incorporating new data.
Deep Learning: is just a subset of machine learning, is an artificial intelligence function that imitates the workings of the human brain in processing data and creating patterns for use in decision making.
Patter Recognition: is the automated recognition of patterns and regularities in data 
Big data: Huge amount of information or data.
Internet Of Things (IoT): It’s a system that allows devices to communicate directly with each other without human intervention.
Let’s talk about the history of Artificial Inteligence.
1950 — Alan Turing creates the “Turing Test” to determine if a machine was really smart. To pass the test, a machine had to be able to deceive a human by making him believe that he was human instead of a computer
1952 — Arthur Samuel writes the first computer program capable of learning. The software was a program that played checkers and improved their game after game.
956 — Martin Minsky and John McCarthy, with the help of Claude Shannon and Nathan Rochester, organize the 1956 Darthmouth conference, considered the event where the field of Artificial Intelligence is born. During the conference, Minsky convinces the attendees to coin the term “Artificial Intelligence” as the name of the new field.
1958 — Frank Rosenblatt designs the Perceptron, the first artificial neural network.
AI-Winter (stagnation time) — [1974–1980]
In the second half of the 70s the field suffered its first “Winter”. Different agencies that finance research in AI cut the funds after many years of high expectations and very little progress.
1979 — Students of Stanford University invent the “Stanford Cart”, a mobile robot capable of moving autonomously through a room avoiding obstacles.
1967 — The “Nearest Neighbor” algorithm is written. This milestone is considered as the birth to the field of pattern recognition in computers.
The explosion of the 80s [1980–1987]
The 80s were marked by the birth of expert systems, based on rules. These were quickly adopted in the corporate sector, which generated a new interest in Machine Learning.
1981 — Gerald Dejong introduces the “Explanation Based Learning” (EBL) concept, where a computer analyzes training data and creates general rules that allow it to discard less important data.
1985 — Terry Sejnowski invents NetTalk, who learns to pronounce words the same way a child would.
Second AI Winter [1987–1993]
At the end of the 80s, and during the first half of the 90s, the second “Winter” of Artificial Intelligence arrived. This time its effects extended for many years and the reputation of the field did not fully recover until the 2000s.
1990s — Work in Machine Learning turns from a knowledge-driven approach to a data-driven one. Scientists begin to create programs that analyze large amounts of data and draw conclusions from the results.
1997 — IBM’s Deep Blue computer defeats world chess champion Gary Kasparov.
Explosion and commercial adoption [2006-Present]
The increase in computing power together with the abundance of available data has re-launched the Machine Learning field. Numerous companies are transforming their businesses towards the data and are incorporating Machine Learning techniques in their processes, products and services to obtain competitive advantages over the competition.
2006 — Geoffrey Hinton coined the term “Deep Learning” to explain new deep Neural Network architectures that are capable of learning much better flatter models.
2011 — The IBM Watson computer defeats its human competitors in the Jeopardy contest which consists of answering questions asked in natural language.
2012 — Jeff Dean, from Google, with the help of Andrew Ng (Stanford University), leads the GoogleBrain project, which develops a Deep Neural Network using the full capacity of Google’s infrastructure to detect patterns in videos and images.
2012 — Geoffrey Hinton leads the winning team of the Computer Vision contest to Imagenet using a Deep Neural Network (RNP). The team won by a wide margin of difference, giving birth to the current explosion of Machine Learning based on RNPs.
2012 — The Google X research laboratory uses GoogleBrain to autonomously analyze YouTube videos and detect those that contain cats.
2014 — Facebook develops DeepFace, an algorithm based on RNPs that is able to recognize people with the same precision as a human being.
2014 — Google buys DeepMind, an English Deep Learning startup that had recently demonstrated the capabilities of Deep Neural Networks with an algorithm capable of playing Atari games simply by viewing the pixels on the screen, just as a person would. The algorithm, after a few hours of training, was able to beat human experts in some of those games.
2015 — Amazon launches its own Machine Learning platform.
2015 — Microsoft creates the “Distributed Machine Learning Toolkit”, which allows the efficient distribution of machine learning problems on multiple computers.
2015 — Elon Musk and Sam Altman, among others, founded the non-profit organization OpenAI, endowing it with 1,000 million dollars in order to ensure that the development of Artificial Intelligence has a positive impact on humanity.
2016 — Google DeepMind wins in the game Go (considered one of the most complicated board games) to the professional player Lee Sedol for 5 games to 1. Go expert players claim that the algorithm was able to make “creative” movements that they have not seen so far.
The collection of “Big Data” and the expansion of the Internet of Things (IoT), has made a perfect environment for new AI applications and services to grow. Applications based on AI are already visible in healthcare diagnostics, targeted treatment, transportation, public safety, service robots, education and entertainment, but will be applied in more fields in the coming years. Together with the Internet, AI changes the way we experience the world and has the potential to be a new engine for economic growth.
Current Uses of AI:
Although artificial intelligence evokes thoughts of science fiction, artificial intelligence already has many uses today, for example.
- Email filtering: Email services use artificial intelligence to filter incoming emails. Users can train their spam filters by marking emails as “spam”.
- Personalization: Online services use artificial intelligence to personalize your experience. Services, like Amazon or Netflix, “learn” from your previous purchases and the purchases of other users in order to recommend relevant content for you.
- Fraud detection: Banks use artificial intelligence to determine if there is strange activity on your account. Unexpected activity, such as foreign transactions, could be flagged by the algorithm.
Machine Learning (ML)
Algorithms are a sequence of instructions used to solve a problem. Algorithms, developed by programmers to instruct computers in new tasks, are the building blocks of the advanced digital world we see today. Computer algorithms organize enormous amounts of data into information and services, based on certain instructions and rules. It’s an important concept to understand, because in machine learning, learning algorithms — not computer programmers — create the rules.
Instead of programming the computer every step of the way, this approach gives the computer instructions that allow it to learn from data without new step-by-step instructions by the programmer. This means computers can be used for new, complicated tasks that could not be manually programmed. Things like photo recognition applications for the visually impaired, or translating pictures into speech.
The basic process of machine learning is to give training data to a learning algorithm. The learning algorithm then generates a new set of rules, based on inferences from the data. This is in essence generating a new algorithm, formally referred to as the machine learning model. By using different training data, the same learning algorithm could be used to generate different models. For example, the same type of learning algorithm could be used to teach the computer how to translate languages or predict the stock market.
Inferring new instructions from data is the core strength of machine learning. It also highlights the critical role of data: the more data available to train the algorithm, the more it learns. In fact, many recent advances in AI have not been due to radical innovations in learning algorithms, but rather by the enormous amount of data enabled by the Internet.
How machines learn:
Although a machine learning model may apply a mix of different techniques, the methods for learning can typically be categorized as three general types:
- Supervised learning: The learning algorithm is given labeled data and the desired output. For example, pictures of dogs labeled “dog” will help the algorithm identify the rules to classify pictures of dogs. Imagine a computer is a child, we are its supervisor (e.g. parent, guardian, or teacher), and we want the child (computer) to learn what a pig looks like. We will show the child several different pictures, some of which are pigs and the rest could be pictures of anything (cats, dogs, etc.). When we see a pig, we shout “pig!” When it’s not a pig, we shout “no, not pig!” After doing this several times with the child, we show them a picture and ask “pig?” and they will correctly (most of the time) say “pig!” or “no, not pig!” depending on what the picture is. That is supervised machine learning.
- Unsupervised learning: The data given to the learning algorithm is unlabeled, and the algorithm is asked to identify patterns in the input data. For example, the recommendation system of an e-commerce website where the learning algorithm discovers similar items often bought together. An unsupervised machine learning algorithm makes use of input data without any labels — in other words, no teacher (label) telling the child (computer) when it is right or when it has made a mistake so that it can self-correct.
- Reinforcement learning: The algorithm interacts with a dynamic environment that provides feedback in terms of rewards and punishments. For example, self-driving cars being rewarded to stay on the road.
Machine learning is not new. Many of the learning algorithms that spurred new interest in the field, such as neural networks , are based on decades old research.  The current growth in AI and machine learning is tied to developments in three important areas:
- Data availability: Just over 3 billion people are online with an estimated 17 billion connected devices or sensors.  That generates a large amount of data which, combined with decreasing costs of data storage, is easily available for use. Machine learning can use this as training data for learning algorithms, developing new rules to perform increasingly complex tasks.
- Computing power: Powerful computers and the ability to connect remote processing power through the Internet make it possible for machine-learning techniques that process enormous amounts of data. 
- Algorithmic innovation: New machine learning techniques, specifically in layered neural networks — also known as “deep learning” — have inspired new services, but is also spurring investments and research in other parts of the field. 
After knowing the history and review some important concepts of AI and ML Grandma, it is time to give you a brief introduction and talk about some of the most used algorithms in machine learning.
Linear Regression: is used to predict (the average) Y from X base on a linear relationship
Logistic Regression: is a mathematical model used in statistics to estimate (guess) the probability of an event occurring having been given some previous data.
Decision Trees: they can be easily visualized so that a human can understand what’s going on. Imagine a flowchart, where each level is a question with a yes or no answer. Eventually an answer will give you a solution to the initial problem.
Support Vector Machines: imagine support vector machine as a “road machine”, which separates the left, right-side cars, buildings, pedestrians and makes the widest lane as possible. And those cars, buildings, really close to the street is the support vectors.
support Vector Machine (the “road machine”) is responsible for finding the decision boundary to separate different classes and maximize the margin.
Margins are the (perpendicular) distances between the line and those dots closest to the line.
K-Nearest Neighbors: The KNN algorithm assumes that similar things exist in close proximity. In other words, similar things are near to each other.
Random Forests: consists of a large number of individual decision trees that operate as an ensemble.
K-Means Clustering: Clustering is a technique for finding similarity groups in a data, called clusters. It attempts to group individuals in a population together by similarity, but not driven by a specific purpose.
Principal Components Analysis: finds the principal components of data. Can be used to reduce the dimensions of a data set.
Deep learning has evolved hand-in-hand with the digital era, which has brought about an explosion of data in all forms and from every region of the world. This data, known simply as big data, is drawn from sources like social media, internet search engines, e-commerce platforms, and online cinemas, among others. This enormous amount of data is readily accessible and can be shared through fintech applications like cloud computing.
A great example of deep learning is Google’s AlphaGo. Google created a computer program with its own neural network that learned to play the abstract board game called Go, which is known for requiring sharp intellect and intuition.
Machine Learning vs Deep Learning
In practical terms, deep learning is just a subset of machine learning. In fact, deep learning technically is machine learning and functions in a similar way (hence why the terms are sometimes loosely interchanged). However, its capabilities are different.
While basic machine learning models do become progressively better at whatever their function is, but they still need some guidance. If an AI algorithm returns an inaccurate prediction, then an engineer has to step in and make adjustments. With a deep learning model, an algorithm can determine on its own if a prediction is accurate or not through its own neural network.
We get it — all of this might still seem complicated. The easiest takeaway for understanding the difference between machine learning and deep learning is to know that deep learning is machine learning.
More specifically, deep learning is considered an evolution of machine learning. It uses a programmable neural network that enables machines to make accurate decisions without help from humans.
Well grandma is all for today, I hope you enjoy this ride and have learned about Artificial Intelligence, Machine Learning and Deep Learning.
 Artificial Intelligence A modern Aproach — Stuart Russell and Peter Norvig
 Neural networks is a computational approach modeled on the human brain.
 The history of neural networks is often described as starting with a seminal paper in 1943 by Warren McCulloch and Walter Pitts on how neurons might work, and where they modeled a simple neural network with electrical circuits.
 Google’s ground breaking experiment “AlphaGo”, the first AI to beat the human champion at the board game Go used approximately 280 GPU cards and 1,920 standard processors. See http://www.economist.com/news/science-andtechnology/21694540-win-or-lose-best-five-battle-contest-another-milestone
 A good example of such progress is the program Libratus, the first AI to beat several of the top human players in no-limit Texas Hold ’Em poker- a game that has been notoriously difficult for an AI to win due to incomplete information about the game state. See https://www.wired.com/2017/02/libratus/