She lived 1815 to 1852. Calculating machines cannot be creative, but only do what program indicates {Lady Lovelace's objection}.
He lived 1912 to 1954 and developed Turing test for intelligence. He developed a code-breaking machine {electronic cryptanalytic machine}, which was the first programmed computer {Colossus computer}. Fixed definite processes {algorithm, Turing} {recursive procedure, Turing} or trial-and-error procedure {heuristic procedure} can solve mathematical problems. Turing machines programmed to perform procedures can solve problems. Universal Turing machines can define all possible operations and solve general problems. Algorithms and heuristics cannot solve some mathematical problems, so machines cannot solve them [Turing, 1950].
He lived 1902 to 1950, studied statistical decision problem, and used minimax [1939].
He lived 1894 to 1964 and studied non-linear problems. He developed animal and machine control and communication theory {cybernetics, Wiener} and feedback-using self-regulating system theory. He helped develop artificial limbs. He studied automata {logical net}, information theory, and principles involved in communication between sources and sinks.
Epistemology
Information has encoding, transmission, and decoding. A possibly noisy channel transmits information. Channel has information capacity. The same information can use different codes, one of which can be optimum.
He lived 1903 to 1957.
In logic, he studied empirical logic, logic with uncertainty, error, and logical-net errors and helped develop quantum logic, with Birkhoff and Mackey.
In computing, he studied linear programming and electronic digital-computer theory and developed first digital computer [1946], called ENIAC. Multiple connections between elements allow system to operate, even if some units fail {multiplexing, Neumann}. Multiple lines can provide multiplexing.
In biology, he studied finite automata as central-nervous-system models.
In geometry, he showed how to use general eigenvalue theory for axiomatic Hilbert spaces and operators.
In game theory, he studied zero-sum games, strategy, Colonel Blotto game, minimax theorem, utility function, prisoner's dilemma, competition, and cooperation. Game theory involves decision-making when conditions are uncertain.
Set theory does not allow sets {paradoxical set, Neumann} to be their own elements {Foundation axiom, Neumann}.
He lived 1915 to 1998 and invented Hamming code [1950] to detect computer-coding errors.
He lived 1934 to ? and invented reversible-computing gates {Fredkin gate} {Conservative Logic Gate}. Mathematical models can reversibly transform into computational models {Fredkin transforms}. Universe computes using discrete and finite quantities {digital mechanics}. The more two alternatives are similar, the harder it is to choose and the less the choice matters {Fredkin paradox, Fredkin}.
He lived 1921 to ?, invented computer programs to prove first-order theorems [1959], and invented infinite series of types. Mathematics is intuitive.
He lived 1928 to ?, studied learning theories {constructionism}, and invented the Logo computer language.
He lived 1923 to ? and wrote ELIZA program to imitate psychologist querying patient.
He studied neural networks.
He studied AI.
They studied AI.
He invented backpropagation learning algorithms.
He invented robots.
He studied neural networks.
He studied quantum computation [1985].
They studied neural networks, with Gregory E. Hinton and R. J. Williams.
He studied computer memory.
He studied computer memory.
To hidden layer, he added units {context layer, Elman} that received a hidden-layer copy and then added back to hidden layer {simple recurrent network, Elman}.
He studied shape from shading in neural networks [1992], with Sidney Lehky.
He invented Mathematica software.
Science does not need laws expressed as mathematical equations. Simple non-linear rules operating on simple units can generate all pattern types and describe all phenomena. Because they can be equivalent to any algorithm, cellular automata can describe all complex processes. Physical systems satisfying differential equations can be cellular automatons, by substituting finite differences and discrete variables for differential equations. His Rule 30 seems to create unpredictable pattern, rather than expected recursiveness. www.stephenwolfram.com/publications/articles/date.html.
Information is only in physical media, which store bits or qubits {information science, Schumacher}. Physical medium can transform and/or transfer information to process information. Output from processing must be verifiable or complete task.
Salesmen want to travel shortest distance among cities, with no path duplication. What is the shortest path {traveling-salesman problem, Kirkpatrick} [2001]? Traveling-salesman problems are NP-complete. Number of possible paths is factorial of number of cities, divided by two, because trips can be in either direction. Tours are vertexes of N-dimensional polygons. Tours that differ by one city are near each other in N-dimensional space. Simulated annealing can find shorter paths but allow longer paths, to avoid local minima. Techniques can find good paths but not necessarily the best.
Outline of Knowledge Database Home Page
Description of Outline of Knowledge Database
Date Modified: 2022.0225