Human-based computation: Difference between revisions

Revision as of 14:36, 29 January 2024

Revision as of 14:37, 29 January 2024

Line 7:

==Early work==

Human-based computation (apart from the [[Computer (job description)|historical meaning of "computer]]") research has its origins in the early work on [[interactive evolutionary computation]] (EC).<ref>{{cite book |url=https://link.springer.com/chapter/10.1007/3-540-61723-X_966 |author=Herdy, Michael |title=Evolution strategies with subjective selection. Basic Concepts of Evolutionary Computation. Volumen 1141, pp. 22-31 |date=1996|pages=22–31 |doi=10.1007/3-540-61723-X_966 |isbn=9783540706687 |access-date=12 May 2022}}</ref> The idea behind interactive evolutionary algorithms has been attributed to [[Richard Dawkins]]; in the Biomorphs software accompanying his book ''[[The Blind Watchmaker]]'' (Dawkins, 1986)<ref>{{cite web|url=https://archive.org/stream/BlindWatchmakerTheRichardDawkins/Blind_Watchmaker_The_-_Richard_Dawkins_djvu.txt|title=''The Blind Watchmaker'' |access-date=12 May 2022|author=Dawkins, Richard}}</ref> the preference of a human experimenter is used to guide the evolution of two-dimensional sets of line segments. In essence, this program asks a human to be the fitness function of an evolutionary algorithm, so that the algorithm can use human visual perception and aesthetic judgment to do something that a normal evolutionary algorithm cannot do. However, it is difficult to get enough evaluations from a single human if we want to evolve more complex shapes. [[Victor Johnston]]<ref>{{cite web|url=http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=5375195.PN.&OS=PN/5375195&RS=PN/5375195 |title=''Method and apparatus for generating composites of human faces'' |access-date=12 May 2022|author=Johnston, Victor}} {{US patent|5375195}}</ref> and [[Karl Sims]]<ref>{{cite web |url=http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=6088510.PN.&OS=PN/6088510&RS=PN/6088510 |title=''Computer system and method for generating and mutating objects by iterative evolution'' |access-date=12 May 2022 |author=Sims, Karl P.}} {{US patent|6088510}}</ref> extended this concept by harnessing the power of many people for fitness evaluation (Caldwell and Johnston, 1991; Sims, 1991). As a result, their programs could evolve beautiful faces and pieces of art appealing to the public. These programs effectively reversed the common interaction between computers and humans. In these programs, the computer is no longer an agent of its user, but instead, a coordinator aggregating efforts of many human evaluators. These and other similar research efforts became the topic of research in aesthetic selection or [[interactive evolutionary computation]] (Takagi, 2001), however the scope of this research was limited to outsourcing evaluation and, as a result, it was not fully exploring the full potential of the outsourcing.

A concept of the automatic [[Turing test]] pioneered by [[Moni Naor]] (1996)<ref>{{cite web |url=http://www.wisdom.weizmann.ac.il/~naor/PAPERS/human_abs.html |author=Naor, Moni |title=Verification of a human in the loop or Identification via the Turing Test |access-date=12 May 2021}}</ref> is another precursor of human-based computation. In Naor's test, the machine can control the access of humans and computers to a service by challenging them with a [[natural language processing]] (NLP) or [[computer vision]] (CV) problem to identify humans among them. The set of problems is chosen in a way that they have no algorithmic solution that is both effective and efficient at the moment. If it existed, such an algorithm could be easily performed by a computer, thus defeating the test. In fact, Moni Naor was modest by calling this an automated Turing test. The [[imitation game]] described by [[Alan Turing]] (1950) didn't propose using CV problems. It was only proposing a specific NLP task, while the Naor test identifies and explores a large [[AI-complete|class]] of problems, not necessarily from the ___domain of NLP, that could be used for the same purpose in both automated and non-automated versions of the test.