AI Glossary - AIC Or Akaike Information Criteria.

 


Akaike Information Criteria or the AIC is a data-driven metric for evaluating several models for the same set of data.

It was calculated by taking into account the loss of accuracy in a model when data-based estimations of the model's parameters were substituted for the true values.

A constant term specified by the real model, -2 times the probability for the data given the model, and a constant multiple (2) of the number of parameters in the model make up the equation for this loss.

The first phrase, involving the unknown true model, may be omitted since it enters as a constant (for a given set of data), leaving two known terms to be assessed.

AIC is the sum of a (negative) measure of the model's mistakes plus a positive penalty for the number of parameters in the model, expressed algebraically.

Increasing the model's complexity will increase the AIC only if the fit (as judged by the data's log-likelihood) improves more than the cost of the new parameters.

A group of competing models may be compared by calculating their AIC values and selecting the model with the lowest AIC value, implying that this model is the most similar to the real model.

Unlike traditional statistical methods, this enables for the comparison of models with no shared parameters.


Related Terms:

Kullback-Liebler information measure, Schwartz Information Criteria



~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.



AI Glossary - Airty.

 



An object's airty is the total number of objects it contains or accepts.



~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.



AI Glossary - AI-QUIC.

 


The underwriting division of American International Groups uses AI-QUIC, a rule-based program.

It automates underwriting activities and is designed to respond fast to changes in underwriting regulations.


Related Terms:

Expert System.


~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.




AI Glossary - Agent CLIPS

 


Agent CLIPS is a CLIPS plugin that enables you to create intelligent agents that can interact with one other on a single system or over the Internet.


Related Terms:

CLIPS.


References:


http://users.aimnet.com/~yilsoft/softwares/agentclips/agentclips.html



~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.



AI Glossary - Agenda Based Systems

 


An agenda or job-list controls the inference process in Agenda Based Systems.

It deconstructs the system into discrete, modular stages.

Each entry in the work list, or task, represents a particular task to be completed throughout the problem-solving process.


Related Terms:


AM, DENDRAL.


~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.


AI Glossary - Advice Taker

 


J. McCarthy suggested a software called Advice Taker to demonstrate common sense and improveable behavior.

Declarative and imperative sentences were used to express the software.

It reasoned through deductive reasoning.

This method was a predecessor to McCarthy and Hayes' Situational Calculus, which they proposed in a 1969 paper in Machine Intelligence.


~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.


AI Glossary - Advanced Scout

 


Advanced Scout  is a customized system built by IBM in the mid-1990s that organizes and interprets data from basketball games using Data Mining methods.


~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.


AI Glossary - ART Or Advanced Reasoning Tool.

 


ART stands for "Advanced Reasoning Tool." 

The Advanced Reasoning Tool (ART) is a knowledge engineering language built on the LISP programming language.

It is a rule-based system that also allows for the modeling of frames and procedures.

Inference Corporation was responsible for its creation.

The same term (ART) is also used to refer to Adaptive Resonance Theory-based approaches.


~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.


AI Glossary - Adjacency Matrix.

 


Adjacency Matrix is an acronym for "adjacency matrix." 

A binary relation over a finite set may be represented using an adjacency matrix.

If the cardinality of set A is n, the adjacency matrix for a relation on A is a nxn binary matrix, with a one for the I j-th element if the connection holds, and a zero otherwise.

The adjacency matrix is used by a variety of route and closure algorithms, either implicitly or explicitly.

If there are ones along the major diagonal, an adjacency matrix is reflexive, and symmetric if the I j-th element equals the j, i-th element for all I j pairings in the matrix.

The weighted adjacency matrix, which substitutes the zeros and ones with and costs, respectively, and utilizes this matrix to determine the shortest distance or minimal cost pathways between the components, is a generalization of this.


Related Terms:


Floyd's Shortest Distance Algorithm, path matrix


AI Glossary - ADE Monitor.

 


ADE Monitor is a CLIPS-based expert system that analyzes patient data for signs of an adverse drug response.

The technology will allow doctors to make changes and will be able to alert the proper authorities as necessary.


Related Terms:


C Language Integrated Production System (CLIPS)


References:


http://www-uk.hpl.hp.com/people/ewc/list-main.html.



~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.


AI Glossary - Additivity And Variance Stabilization (AVAS).

 



AVAS stands for Additivity and Variance Stabilization and is a smooth regression model version of the ACE approach.

It incorporates a variance stabilizing transform into the ACE approach, removing many of ACE's problems with smooth connection estimation.


Related Terms:

ACE



~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.


AI Glossary - Additive Models.

 


A modeling approach that predicts the output variable using weighted linear sums of the presumably altered input variables, but excludes phrases like cross-products that are dependent on more than one predictor variable.

Boosting and Generalized Additive Models are two examples of machine learning methods that employ additive models (GAMs).


Related Terms:


Boosting, Generalized Additive Models.



~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.


AI Glossary - Adaptive Vector Quantization

 


Adaptive Vector Quantization is a neural network technique in which the input vectors form a state space and the network quantizes those vectors into a smaller number of ideal vectors or regions.

The network adapts the placement (and quantity) of these vectors to the data as it "learns." 



~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.


AI Glossary - Adaptive Resonance Theory (ART).


 


Adaptive Resonance Theory (ART) is a kind of neural network based on neurophysiologic theories.

Stephen Grossberg came up with the idea in 1976.

For prediction, ART models use a hidden layer of ideal instances.

If an input case is sufficiently similar to an existing case, it "resonates" with it, and the ideal case is modified to include it.

A new ideal scenario is introduced if this is not the case.

ARTs are sometimes shown as having two layers, known as the F1 and F2 layers.

The matching is done by the F1 layer, and the outcome is chosen by the F2 layer.

It's a cluster analysis technique.



Internet References:


http://www.wi.leidenuniv.nl/art/

ftp:://ftp.sas.com/pub/neural/FAQ2.html


~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.


Artificial Intelligence - Who Is Aaron Sloman?

 




Aaron Sloman (1936–) is a renowned artificial intelligence and cognitive science philosopher.

He is a global expert in the evolution of biological information processing, an area of study that seeks to understand how animal species have acquired cognitive levels that surpass technology.

He's been debating if evolution was the first blind mathematician and whether weaver birds are actually capable of recursion in recent years (dividing a problem into parts to conquer it).

His present Meta-Morphogenesis Project is based on an idea by Alan Turing (1912–1954), who claimed that although computers could do mathematical brilliance, only brains could perform mathematical intuition.

According to Sloman, not every aspect of the cosmos, including the human brain, can be represented in a sufficiently massive digital computer because of this.

This assertion clearly contradicts digital physics, which claims that the universe may be characterized as a simulation running on a sufficiently big and fast general-purpose computer that calculates the cosmos's development.

Sloman proposes that the universe has developed its own biological building kits for creating and deriving other—different and more sophisticated—construction kits, similar to how scientists have evolved, accumulated, and applied increasingly complex mathematical knowledge via mathematics.

He refers to this concept as the Self-Informing Universe, and suggests that scientists build a multi-membrane Super-Turing machine that runs on subneural biological chemistry.

Sloman was born to Jewish Lithuanian immigrants in Southern Rhodesia (now Zimbabwe).

At the University of Cape Town, he got a bachelor's degree in Mathematics and Physics.

He was awarded a Rhodes Scholarship and earned his PhD in philosophy from Oxford University, where he defended Immanuel Kant's mathematical concepts.

He saw that artificial intelligence had promise as the way forward in philosophical understanding of the mind as a visiting scholar at Edinburgh University in the early 1970s.

He said that using Kant's recommendations as a starting point, a workable robotic toy baby could be created, which would eventually develop in intellect and become a mathematician on par with Archimedes or Zeno.

He was one of the first scholars to refute John McCarthy's claim that a computer program capable of operating intelligently in the real world must use structured, logic-based ideas.

Sloman was one of the founding members of the University of Sussex School of Cognitive and Computer Sciences.

There, he collaborated with Margaret Boden and Max Clowes to advance artificial intelligence instruction and research.

This effort resulted in the commercialization of the widely used Poplog AI teaching system.

Sloman's The Computer Revolution in Philosophy (1978) is famous for being one of the first to recognize that metaphors from the realm of computers (for example, the brain as a data storage device and thinking as a collection of tools) will dramatically alter how we think about ourselves.

The epilogue of the book contains observations on the near impossibility of AI sparking the Singularity and the likelihood of a human Society for the Liberation of Robots to address possible future brutal treatment of intelligent machines.

Sloman held the Artificial Intelligence and Cognitive Science chair in the School of Computer Science at the University of Birmingham until his formal retirement in 2002.

He is a member of the Alan Turing Institute and the Association for the Advancement of Artificial Intelligence.


~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


You may also want to read more about Artificial Intelligence here.



See also: 


Superintelligence; Turing, Alan.


References & Further Reading:


Sloman, Aaron. 1962. “Knowing and Understanding: Relations Between Meaning and Truth, Meaning and Necessary Truth, Meaning and Synthetic Necessary Truth.” D. Phil., Oxford University.

Sloman, Aaron. 1971. “Interactions between Philosophy and AI: The Role of Intuition and Non-Logical Reasoning in Intelligence.” Artificial Intelligence 2: 209–25.

Sloman, Aaron. 1978. The Computer Revolution in Philosophy: Philosophy, Science, and Models of Mind. Terrace, Hassocks, Sussex, UK: Harvester Press.

Sloman, Aaron. 1990. “Notes on Consciousness.” AISB Quarterly 72: 8–14.

Sloman, Aaron. 2018. “Can Digital Computers Support Ancient Mathematical Conscious￾ness?” Information 9, no. 5: 111.



AI Terms Glossary - Adaptive Fuzzy Associative Memory (AFAM).

 


A fuzzy associative memory that can adapt to changing input over time.



~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.







AI Terms Glossary - Adaptive



A term used to describe systems that can learn or adjust from data in use, such as neural networks or other dynamic control systems.



~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.







AI Terms Glossary - ADABOOST.MH

 


ADABOOST.MH is a multi-class and multi-label data extension of the ADABOOST algorithm.


See Also: 


multi-class, multi-label.


~ Jai Krishna Ponnappan

Find Jai on Twitter | LinkedIn | Instagram


Be sure to refer to the complete & active AI Terms Glossary here.

You may also want to read more about Artificial Intelligence here.







Analog Space Missions: Earth-Bound Training for Cosmic Exploration

What are Analog Space Missions? Analog space missions are a unique approach to space exploration, involving the simulation of extraterrestri...