Hybrid intelligence

from Wikipedia, the free encyclopedia

Hybrid intelligence is a new name for general artificial intelligence by combining human intelligence and the intelligence of machines . The basis is the attempt to combine the strengths of the complementary heterogeneous intelligences. Hybrid intelligence is defined as the ability to achieve complex goals through the interaction of human and artificial intelligence in order to achieve better results together and to learn from each other. Research on hybrid intelligence is an interdisciplinary field that Computer Science , Information Systems Research , Neuroscience , Cognitive Psychologyand industrial engineering .

Aspects of Hybrid Intelligence

intelligence

In the various sciences of psychology, cognitive science, neuroscience, human behavior, education or computer science, there are different definitions and measures for intelligence (e.g. social, logical, spatial, musical, etc.). An inclusive, high-level definition is used for hybrid intelligence, which generally defines intelligence as the ability to achieve complex goals, learn, reason, and interact with the environment .

Human intelligence

The sub-category of Human Intelligence ascribed to humans is defined by the ability to learn, draw conclusions and interact with the environment based on their knowledge. It allows people to adapt to changing environmental conditions and to achieve their goals. The Wechsler definition of human intelligence describes intelligence as a composite or comprehensive ability of an individual to act purposefully, to think sensibly and to interact with his environment.

Collective intelligence

Collective intelligence is also attributed to humans. Collective intelligence refers to "groups of individuals who work collectively together in a way that appears intelligent". Although the term “individuals” leaves room for interpretation, scientists in this area use the concept of the wisdom of the masses and thus a combined intelligence of the individual individuals. The concept describes that an average group of people achieves better results than a single person from this group or a single expert because errors are reduced and knowledge is brought together.

Artificial intelligence

The sub-category of artificial intelligence is assigned to machines. The term refers to systems that have "activities that we associate with human thinking, such as decision-making, problem-solving, and learning." Although there are different definitions, it is generally required that the machine can achieve complex goals. This includes facets such as the processing of language, the perception of objects, the storage of knowledge and the application of this knowledge to solve problems, as well as machine learning to adapt to and work in new environments. There are different approaches to achieving AI. These are more or less strongly connected with understanding and increasing intelligence. For example, cognitive computing aims "at the development of a coherent, unified, universal mechanism inspired by the abilities of the mind and the implementation of a unified computational theory of the mind." Therefore, scientists want to develop machines that "learn and think like humans."

collective

Hybrid intelligence does not take into account that tasks are performed collectively. This means that activities of each part are carried out in an interdependent manner, although this is not always necessary in order to achieve a goal.

Better results

This means that a system achieves results that none of the actors involved would achieve alone in the same task.

Continuous learning

A central aspect is that over time, the socio-technical system as a whole, but also the individual components (e.g. people and machines ) improve through the experience gained in solving a task.

Applications

Web links

Individual evidence

  1. ↑ Ece Kamar: Directions in Hybrid Intelligence: Complementing AI Systems with Human Intelligence .
  2. ↑ Dominik Dellermann, Philip Ebel, Matthias Söllner, Jan Marco Leimeister: Hybrid Intelligence . In: Business & Information Systems Engineering . tape 61 , no. 5 , October 2019, ISSN  2363-7005 , p. 637–643 , doi : 10.1007 / s12599-019-00595-2 ( springer.com [accessed December 16, 2019]).
  3. ↑ Linda S. Gottfredson: Why g matters: The complexity of everyday life . In: Intelligence . tape 24 , no. 1 , January 1997, ISSN  0160-2896 , p. 79-132 , doi : 10.1016 / s0160-2896 (97) 90014-3 .
  4. ^ C. Molz, R. Schulze, U. Schroeders, O. Wilhelm: TBS-TK reviews . In: Psychological Rundschau . tape 61 , no. 4 , October 2010, ISSN  0033-3042 , p. 229–230 , doi : 10.1026 / 0033-3042 / a000042 ( hogrefe.com [accessed May 30, 2018]).
  5. ↑ a b Malone, Thomas W .: Handbook of Collective Intelligence. OCLC 928998019 .
  6. ^ Anita Williams Woolley, Christopher F. Chabris, Alex Pentland, Nada Hashmi, Thomas W. Malone: Evidence for a Collective Intelligence Factor in the Performance of Human Groups . In: Science . tape 330 , no. 6004 , October 29, 2010, ISSN  0036-8075 , p. 686–688 , doi : 10.1126 / science.1193147 , PMID 20929725 ( sciencemag.org [accessed May 30, 2018]).
  7. ↑ Dellermann, D .; Lipusch, N. & Ebel, P .: Heading for new shores: Crowdsourcing for entrepreneurial opportunity creation . In: European Conference of Information Systems (ECIS) . tape 2018 .
  8. ↑ Bellman, Richard, 1920-1984 .: An introduction to artificial intelligence: can computers think? Boyd & Fraser Pub. Co, San Francisco 1978, ISBN 0-87835-066-7 .
  9. ↑ Norvig, Peter 1956-: Artificial intelligence a modern approach . Third ed. Pearson, Boston 2016, OCLC 945899984 .
  10. ↑ Modha, DS, Ananthanarayanan, R., Esser, SK, Ndirango, A., Sherbondy, AJ, & Singh, R .: Cognitive computing . In: Communications of the ACM . tape 54 , no. 8 , p. 62-71 .
  11. ↑ Brenden M. Lake, Tomer D. Ullman, Joshua B. Tenenbaum, Samuel J. Gershman: Building machines that learn and think like people . In: Behavioral and Brain Sciences . tape 40 , 2017, ISSN  0140-525X , doi : 10.1017 / S0140525X16001837 ( cambridge.org [accessed May 30, 2018]).
  12. ↑ Dominik Dellermann, Nikolaus Lipusch, Philipp Ebel, Jan Marco Leimeister: Design principles for a hybrid intelligence decision support system for business model validation . In: Electronic Markets . tape 29 , no. 3 , September 2019, ISSN  1019-6781 , p. 423-441 , doi : 10.1007 / s12525-018-0309-2 ( springer.com [accessed December 16, 2019]).
  13. ↑ Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Andrei A. Rusu, Joel Veness: Human-level control through deep reinforcement learning . In: Nature . tape 518 , no. 7540 , February 2015, ISSN  0028-0836 , p. 529-533 , doi : 10.1038 / nature14236 .
  14. ↑ Dominik Dellermann, Nikolaus Lipusch, Philipp Ebel, Karl Michael Popp, JM Leimeister: Finding the Unicorn: Predicting Early Stage Startup Success Through a Hybrid Intelligence Method . ID 3159123. Social Science Research Network, Rochester, NY December 10, 2017, doi : 10.2139 / ssrn.3159123 ( ssrn.com [accessed May 30, 2018]).
  15. ↑ Dellermann, Dominik; Lipusch, Nikolaus & Li, Mahei: Combining Humans and Machine Learning: A Novel Approach for Evaluating Crowdsourcing Contributions in Idea Contests. In: Multi-Conference Business Informatics (MKWI) . tape 2018 ( unisg.ch [PDF]).
  16. ↑ Joseph Chee Chang, Saleema Amershi, Ece Kamar: Revolt: Collaborative crowdsourcing for labeling machine learning datasets . ACM, 2017, ISBN 978-1-4503-4655-9 , pp. 2334–2346 , doi : 10.1145 / 3025453.3026044 ( acm.org [accessed May 30, 2018]).