Artificial Intelligence Sp23

Agent

a computer program that acts on behalf of a user or other program (also called bots).

Algorithm

a step-by-step set of instructions to performing a specific task that will lead to a desired result when done correctly. They have a definitive start, end, and number of steps in between

Analogical reasoning

compares the similarities between a new concept and something already understood to draw conclusions as to what is likely to be true about the new concept.

Artificial General Intelligence

machines that can figure stuff out on their own

Artificial Intelligence

the ability for computers to make decisions and act as a human would

Artificial neural network (ANN)

also known as a neural network) is a computing system that attempts to mimic the human brain, with layers of connected "neurons" sending information to each other.

Autonomous

the ability to act independently of a ruling body. In AI, a machine or vehicle is referred to as autonomous if it doesn't require input from a human operator to function properly

Backpropagation

- Short for "backward propagation of errors" back propagation is a way of training neural networks based on a known, desired output for specific sample case

Big data

- a term for data sets that are so large or complex that traditional data processing applications are inadequate to deal with them

Case-based learning (CBR)

An approach to knowledge-based problem solving that uses the solutions of a past, similar problem (case) to solve an existing problem

Cognitive computing

software that succeeds at specific tasks previously thought to be limited to human cognition.

Computer vision

the field of A.I. concerned with teaching machines how to interpret the visual world

Dark data

is operational data that is not used. It is captured and stored, but not analyzed, providing no tactical value or insights. According to Lucidworks, we create 7.5 septillion (7,700,000,000,000,000,000,000) gigabytes of data worldwide every day. Businesses typically analyze approximately 10% of the data they collect

Data crunching

The automated analysis of vast amounts of data originating from Big Data. Once imported into a system, the data is sorted, structured, processed informed decisions.

Data mining

the computational process of discovering patterns in large data sets

Deep learning

the application of artificial neural networks (ANNs) to learning tasks that contain more than one hidden layer.

Expert systems

AI paradigm based on representations of human, domain-specific knowledge to replicate specific aspects of human decision making

Embodied A.I.

A fancy way of saying "robots with A.I. capabilities."

Forward chaining

A situation where an AI system must work "forward" from a problem to find a solution. Using a rule-based system, the AI would determine which "if" rules it would apply to the problem

Human-computer interaction

commonly referred to as (HCI) researches the design and use of computer technology, focused on the interfaces between people (users) and computers.

Explainable A.I. (X.A.I.)

A.I. that can tell or show its human operators how it came to its conclusions

Natural language processing

(NLP) is a way for computers to analyze and understand human language.

Few-shot learning

One-shot and few-shot learning try to create a system that can be taught to do something with far less training. It's similar to how toddlers might learn a new concept or task

Optical character recognition

(OCR) the mechanical or electronic conversion of images of typed, handwritten or printed text into machine-encoded text, whether from a scanned document, a photo of a document, a scene-photo or from subtitle text superimposed on an image.

Planning

A branch of AI dealing with planned sequences or strategies to be performed by an AI-powered machine. Things such as actions to take, variable to account for, and duration of performance are accounted for

Pruining

The use of a search algorithm to cut off undesirable solutions to a problem in an AI system. It reduces the number of decisions that can be made by the AI system

Machine learning

(ML) is a type of artificial intelligence in which a machine is trained to learn from past experiences and make decisions when exposed to new information without being explicitly programmed to do so.

Reinforcement learning

A process where machines learn to do a new task like humans do — through a system of rewards and punishments — starting as a novice and improving with practice and feedback.

Strong AI

functions like a human mind. It can think, reason, understand, and act as a human should

Supervised learning

A technique that teaches a machine-learning algorithm to solve a specific task using data that has been carefully labeled by a human. Everyday examples include most weather prediction and spam detection.

Symbolic reasoning systems

AI paradigm based on orchestration of complex logical structures (GOFAI)

Transfer learning

This method tries to take training data used for one thing and reused it for a new set of tasks, without having to retrain the system from scratch.

TURING test

developed by Alan Turing (the father of artificial intelligence) in the 1950s to test a machine's ability to exhibit behavior indistinguishable from that of a human

Unsupervised learning

An approach that gives A.I. unlabeled data and has to make sense of it without any instruction. In essence, it is when machines "teach themselves."

Watson

the world's first cognitive computing system

Weak A.I.

can do just one thing at a time, like play chess or recognize breeds of cats. Appears to think and understand, but it can only do what it is programmed to do. It is bound by rules and is not conscious.