Artificial Intelligence (AI)

What is Deep Blue

A chess-playing computer developed by IBM, won its first game against a world champion on Feb. 10 1996, when it defeated Garry Kasparov in game one of a 6 game match. However, Kasparov won 3 and drew 2 of the following 5 games, defeating Deep Blue 4-2. Deep Blue was then heavily upgraded, and played Kasparov again in May 1997 winning the 6 game rematch 3 1/2-2 1/2 and becoming the first computer system to defeat a reigning world champion in a match under standard chess tournament time controls.

What is AlphaGo Zero

Alpha Go's little brother, that defeated AlphaGo 100-0. Zero was not taught how to play, it learned the best moves over time, simply by playing millions of games against itself. This is an example of reinforcement learning. It learned from scratch. DeepMind co-founder and CEO Demis Hassabis said the program was so powerful because it was "no longer constrained by the limits of human knowledge."

Information about AlphaGo

AlphaGo uses an AI technique called deeplearning. It spent several months in studying 30 million games played by humans and trained itself.

Applications

Any profession within any industry where the amount of information being produced has surpassed the ability of the humans in those professions to consume it.

Examples of Voice Recognition

Deepmind WaveNet

Deduction, Reasoning, Problem Solving

Develop algorithms that imitated the step-by-step reasoning that humans use when they solve puzzles or make logical deductions. With uncertain or incomplete information, probability and economics are needed. For difficult problems, most of these algorithms can require enormous computational resources. The search for more efficient problem-solving algorithms is a high priority for AI research.

What is Motion Planning?

Figuring out how to get there.

What is Path Planning?

Going from one point in space to another point.

Examples of AI

Google Assistant Amazon Echo

Google's Deepmind

Google created AplhaGo, the first computer program to ever beat a professional, human player. AlphaGo won 5-0 in a formal match on Oct. 2015, against the reigning European Champ, Fan Hui, becoming the first program to ever beat a professional Go player. In March 2016 AlphaGo wont 4-1 against Lee Sedol, the top Go player in the world.

Examples of Machine Learning

Google's algorithm recognizes that you searched for something a couple of seconds after searching something else, and it keeps this in mind for future users who make a similar typing mistake. (Typing "phine" instead of "phone")

Define Sub-symbolic Problem Solving

Human solve most of their problems using fast, intuitive judgments rather than the conscious, step-by-step deduction.

Prisoner's Dilemma

If A and B each betray the other, each of them serves 2 years in prison. If A betrays B but B remains silent, A will be set free and B will serve 3 years in prison. If A and B both remain silent, both of them will only serve 1 year in prison.

Planning

Intelligent agents must be able to set goals and achieve them. They need a way to visualize the future. They must have a representation of the state of the world and be able to make predictions about how their actions will change it and be able to make choices that maximize the utility (or "value") of the available choices

Application Examples

Intrusion Detection Online and telephone customer service Fraud Prevention Automobile Robots Healthcare Toys

Knowledge Representation

Many of the problems machines are expected to solve require extensive knowledge about the world: objects, properties, categories, relations between objects, situations, events, states, time, cause & effects, knowledge about knowledge (what we know about what other people know)

What did AlphaGo Zero develop?

It developed intriguing new strategies of its own and had "genuine moments of creativity."

AlphaGo Zero- Learning from Scratch

It does not need human knowledge. Needs only short amount of time (40 days played 29 million games) and limited computing resources (only 4 TPUs- tensor processing units). Used reinforced learning.

Natural Language Processing

It enables machines the ability to read and understand the languages that humans speak

What is Supervised Learning

It includes both classification and numerical regression

What is Intelligent Agent

It is a system that perceives its environment and takes actions that maximize its chances of success.

What are Chatbots/Chatterbot/Chatter Robot

It is a type of conversational agent, a computer program designed to simulate an intelligent conversation with 1 or more human users via auditory or textual methods.

What is AI?

It is intelligence exhibited by machines It creates machines software that are capable of intelligent behavior.

What is the Central Belief

It is that human intelligence can be so precisely described that a machine can be made to simulate it.

What is Unsupervised Learning

It is the ability to find patterns in a stream of input

Motion and Manipulation

It is the ability to handle tasks like object manipulation and navigation.

What is Perception

It is the ability to use input from sensors to deduce aspects of the world.

What is Regression

It is the attempt to produce a function that describes the relationship between inputs and outputs and predicts how the outputs should change as inputs change.

What is Machine Learning

It is the study of computer algorithms that improve automatically through experience.

How is Classification used?

It is used to determine what category something belongs in, after seeing a number of examples of things from several categories.

What does Combinatorial Explosion mean?

It means, the amount of memory or computer time required becomes astronomical when the problem goes beyond a certain size.

What type of learning did AlphaGo Zero use

It uses "Tabula Rasa" or blank slate learning, in which the program becomes its own teacher, playing games against itself and improving a little each time. It needs no human knowledge, data or any intervention.

What is Localization?

Knowing where you are, or finding out where other things are.

What is Mapping?

Learning what is around you, building a map of the environment.

Neural Network

Neural Network based AI attempts to simulate the structures inside the brain.

Long-Term Goals

Social Intelligence- the machine should interpret the emotional stat of humans and adapt its behavior to them, giving an appropriate response for those emotions. Creativity- generate outputs that can be considered creative General Intelligence- known as strong AI, combining all the skills above and exceeding human abilities at most or all of them.

What are the capabilities of AI?

Successfully understanding human speech Competing at a high level in strategic game systems Self-driving cars Interpreting complex data

Statistical Approach

Such as Bayesian Networks, to AI mimic the probabilistic nature of the human ability to guess.

One Agent Planning

The agent can assume that it is the only thing acting on the world and it can be certain what the consequences of its actions may be

Multi-Agent Planning

The agent is not the only actor, it must periodically ascertain whether the world matches its predictions and it must change its plan as this becomes necessary. requiring the agent to reason under uncertainty. Uses the cooperation and competition of many agents to achieve a given goal.

Reinforcement Learning

The agent is rewarded for good response and punished for bad ones. The agent uses this sequence of rewards and punishments to form a strategy for operation in its problem space.

What is Watson

The smartest machine ever built On January 14th 2011, IBM Watson defeated 2 legendary Jeopardy champions, Ken Jennings and Brad Rutter.

What are Omnipresent Devices?

They work in the background, predicting your needs, processing your requests, and offering neatly parceled answers to your questions.

AI Effect

When an AI technique reaches mainstream use, it is no longer considered artificial intelligence.