398:" to mutate and preferentially replicate high-scoring AI systems, similar to how animals evolved to innately desire certain goals such as finding food. Some AI systems, such as nearest-neighbor, instead of reason by analogy, these systems are not generally given goals, except to the degree that goals are implicit in their training data. Such systems can still be benchmarked if the non-goal system is framed as a system whose "goal" is to accomplish its narrow classification task.
604:
819:
654:
625:
576:
251:
545:
375:, 0 otherwise") or complex ("Perform actions mathematically similar to ones that succeeded in the past"). The "goal function" encapsulates all of the goals the agent is driven to act on; in the case of rational agents, the function also encapsulates the acceptable trade-offs between accomplishing conflicting goals. (Terminology varies; for example, some agents seek to maximize or minimize a "
89:
724:
Intelligent agents can be organized hierarchically into multiple "sub-agents". Intelligent sub-agents process and perform lower-level functions. Taken together, the intelligent agent and sub-agents create a complete system that can accomplish difficult tasks or goals with behaviors and responses that
644:
A rational utility-based agent chooses the action that maximizes the expected utility of the action outcomes - that is, what the agent expects to derive, on average, given the probabilities and utilities of each outcome. A utility-based agent has to model and keep track of its environment, tasks that
426:
can generate intelligent agents that appear to act in ways intended to maximize a "reward function". Sometimes, rather than setting the reward function to be directly equal to the desired benchmark evaluation function, machine learning programmers will use reward shaping to initially give the machine
401:
Systems that are not traditionally considered agents, such as knowledge-representation systems, are sometimes subsumed into the paradigm by framing them as agents that have a goal of (for example) answering questions as accurately as possible; the concept of an "action" is here extended to encompass
728:
Generally, an agent can be constructed by separating the body into the sensors and actuators, and so that it operates with a complex perception system that takes the description of the world as input for a controller and outputs commands to the actuator. However, a hierarchy of controller layers is
661:
Learning has the advantage of allowing agents to initially operate in unknown environments and become more competent than their initial knowledge alone might allow. The most important distinction is between the "learning element", responsible for making improvements, and the "performance element",
338:
More importantly, it has a number of practical advantages that have helped move AI research forward. It provides a reliable and scientific way to test programs; researchers can directly compare or even combine different approaches to isolated problems, by asking which agent is best at maximizing a
435:
chess had a simple objective function; each win counted as +1 point, and each loss counted as -1 point. An objective function for a self-driving car would have to be more complicated. Evolutionary computing can evolve intelligent agents that appear to act in ways intended to maximize a "fitness
583:
A model-based agent can handle partially observable environments. Its current state is stored inside the agent maintaining some kind of structure that describes the part of the world which cannot be seen. This knowledge about "how the world works" is called a model of the world, hence the name
611:
Goal-based agents further expand on the capabilities of the model-based agents, by using "goal" information. Goal information describes situations that are desirable. This provides the agent a way to choose among multiple possibilities, selecting the one which reaches a goal state. Search and
867:. It simulates traffic interactions between human drivers, pedestrians and automated vehicles. People's behavior is imitated by artificial agents based on data of real human behavior. The basic idea of using agent-based modeling to understand self-driving cars was discussed as early as 2003.
236:
and
Haenlein define artificial intelligence as "a system's ability to correctly interpret external data, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation". This definition is closely related to that of an intelligent agent.
665:
The learning element uses feedback from the "critic" on how the agent is doing and determines how the performance element, or "actor", should be modified to do better in the future. The performance element, previously considered the entire agent, takes in percepts and decides on actions.
591:
that depends on the percept history and thereby reflects at least some of the unobserved aspects of the current state. Percept history and impact of action on the environment can be determined by using the internal model. It then chooses an action in the same way as reflex agent.
217:
Padgham & Winikoff (2005) agree that an intelligent agent is situated in an environment and responds in a timely (though not necessarily real-time) manner to changes in the environment. However, intelligent agents must also proactively pursue goals in a flexible and
96:
Leading AI textbooks define "artificial intelligence" as the "study and design of intelligent agents", a definition that considers goal-directed behavior to be the essence of intelligence. Goal-directed agents are also described using a term borrowed from
519:
We use the term percept to refer to the agent's perceptional inputs at any given instant. In the following figures, an agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators.
406:
of the 2010s, an "encoder"/"generator" component attempts to mimic and improvise human text composition. The generator is attempting to maximize a function encapsulating how well it can fool an antagonistic "predictor"/"discriminator" component.
858:
Hallerbach et al. discussed the application of agent-based approaches for the development and validation of automated driving systems via a digital twin of the vehicle-under-test and microscopic traffic simulation based on independent agents.
402:
the "act" of giving an answer to a question. As an additional extension, mimicry-driven systems can be framed as agents who are optimizing a "goal function" based on how closely the IA succeeds in mimicking the desired behavior. In the
563:
This agent function only succeeds when the environment is fully observable. Some reflex agents can also contain information on their current state which allows them to disregard conditions whose actuators are already triggered.
370:
An agent that is assigned an explicit "goal function" is considered more intelligent if it consistently takes actions that successfully maximize its programmed goal function. The goal can be simple ("1 if the IA wins a game of
447:. In the real world, an IA is constrained by finite time and hardware resources, and scientists compete to produce algorithms that can achieve progressively higher scores on benchmark tests with existing hardware.
459:
f (called the "agent function") which maps every possible percepts sequence to a possible action the agent can perform or to a coefficient, feedback element, function or constant that affects eventual actions:
1356:
Adams, Sam; Arel, Itmar; Bach, Joscha; Coop, Robert; Furlan, Rod; Goertzel, Ben; Hall, J. Storrs; Samsonovich, Alexei; Scheutz, Matthias; Schlesinger, Matthew; Shapiro, Stuart C.; Sowa, John (15 March 2012).
335:"). It also doesn't attempt to draw a sharp dividing line between behaviors that are "intelligent" and behaviors that are "unintelligent"—programs need only be measured in terms of their objective function.
632:
Goal-based agents only distinguish between goal states and non-goal states. It is also possible to define a measure of how desirable a particular state is. This measure can be obtained through the use of a
567:
Infinite loops are often unavoidable for simple reflex agents operating in partially observable environments. If the agent can randomize its actions, it may be possible to escape from infinite loops.
693:
Layered architectures – in which decision-making is realized via various software layers, each of which is more or less explicitly reasoning about the environment at different levels of abstraction
1177:
Kaplan, Andreas; Haenlein, Michael (1 January 2019). "Siri, Siri, in my hand: Who's the fairest in the land? On the interpretations, illustrations, and implications of artificial intelligence".
690:
Belief-desire-intention agents – in which decision making depends upon the manipulation of data structures representing the beliefs, desires, and intentions of the agent; and finally,
500:
1308:
Andrew Y. Ng, Daishi Harada, and Stuart
Russell. "Policy invariance under reward transformations: Theory and application to reward shaping." In ICML, vol. 99, pp. 278-287. 1999.
108:
An agent has an "objective function" that encapsulates all the IA's goals. Such an agent is designed to create and execute whatever plan will, upon completion, maximize the
160:
Intelligent agents are often described schematically as an abstract functional system similar to a computer program. Abstract descriptions of intelligent agents are called
1507:
Stefano
Albrecht and Peter Stone (2018). Autonomous Agents Modelling Other Agents: A Comprehensive Survey and Open Problems. Artificial Intelligence, Vol. 258, pp. 66-95.
1606:"1.3 Agents Situated in Environments‣ Chapter 2 Agent Architectures and Hierarchical Control‣ Artificial Intelligence: Foundations of Computational Agents, 2nd Edition"
669:
The last component of the learning agent is the "problem generator". It is responsible for suggesting actions that will lead to new and informative experiences.
1631:
1709:
641:
of different world states according to how well they satisfied the agent's goals. The term utility can be used to describe how "happy" the agent is.
741:". Some 20th-century definitions characterize an agent as a program that aids a user or that acts on behalf of a user. These examples are known as
1330:
1247:
Lindenbaum, M., Markovitch, S., & Rusakov, D. (2004). Selective sampling for nearest neighbor classifiers. Machine learning, 54(2), 125–152.
227:
431:
stated in 2018, "Most of the learning algorithms that people have come up with essentially consist of minimizing some objective function."
315:, it does not refer to human intelligence in any way. Thus, there is no need to discuss if it is "real" vs "simulated" intelligence (i.e.,
268:
882:
745:, and sometimes an "intelligent software agent" (that is, a software agent with intelligence) is referred to as an "intelligent agent".
1284:
1257:
1077:
189:
829:
729:
often necessary to balance the immediate reaction desired for low-level tasks and the slow reasoning about complex, high-level goals.
1870:
1853:
394:" that encourages some types of behavior and punishes others. Alternatively, an evolutionary system can induce goals by using a "
1167:
Lin
Padgham and Michael Winikoff. Developing intelligent agent systems: A practical guide. Vol. 13. John Wiley & Sons, 2005.
1655:
Burgin, Mark, and
Gordana Dodig-Crnkovic. "A systematic approach to artificial agents." arXiv preprint arXiv:0902.3513 (2009).
197:"Anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators"
1896:
1439:
1391:
205:"An agent that acts so as to maximize the expected value of a performance measure based on past experience and knowledge."
128:
505:
Agent function is an abstract concept as it could incorporate various principles of decision making like calculation of
1814:
1842:
1770:
638:
613:
553:
298:
280:
1927:
924:
687:
Reactive agents – in which decision making is implemented in some form of direct mapping from situation to action.
938:
616:
are the subfields of artificial intelligence devoted to finding action sequences that achieve the agent's goals.
403:
1521:
1058:
637:
which maps a state to a measure of the utility of the state. A more general performance measure should allow a
261:
1734:
Connors, J.; Graham, S.; Mailloux, L. (2018). "Cyber
Synthetic Modeling for Vehicle-to-Vehicle Applications".
466:
172:
is designed to function in the absence of human intervention. Intelligent agents are also closely related to
1014:
The
Padgham & Winikoff definition explicitly covers only social agents that interact with other agents.
839:
801:
311:
Philosophically, this definition of artificial intelligence avoids several lines of criticism. Unlike the
1320:. Architects of Intelligence: The truth about AI from the people building it. Packt Publishing Ltd, 2018.
595:
An agent may also use models to describe and predict the behaviors of other agents in the environment.
684:
Logic-based agents – in which the decision about what action to perform is made via logical deduction.
339:
given "goal function". It also gives them a common language to communicate with other fields—such as
943:
415:
340:
116:
agent has a "reward function" that allows the programmers to shape the IA's desired behavior, and an
58:
958:
953:
276:
27:
973:
892:
645:
have involved a great deal of research on perception, representation, reasoning, and learning.
536:
group agents into five classes based on their degree of perceived intelligence and capability:
456:
423:
419:
387:
316:
117:
113:
1679:"Simulation-Based Identification of Critical Scenarios for Cooperative and Automated Vehicles"
967:
1803:
The Master
Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World
1755:
Proceedings of the 2003 IEEE International
Conference on Intelligent Transportation Systems
1558:
1317:
877:
272:
150:
20:
8:
948:
835:
768:
1562:
1875:
1776:
1194:
1122:
963:
921:– the ability for agents to search heterogeneous data sources using a single vocabulary
719:
46:
1912:
1892:
1879:
1838:
1824:
1810:
1780:
1766:
1678:
1586:
1478:
1473:
1456:
1435:
1198:
984:
738:
332:
154:
144:
132:
124:
414:
systems often accept an explicit goal function, the paradigm can also be applied to
1865:
1758:
1690:
1605:
1576:
1571:
1566:
1546:
1468:
1370:
1186:
1126:
1114:
918:
887:
864:
863:
has created a multi-agent simulation environment, Carcraft, to test algorithms for
794:
788:
764:
737:"Intelligent agent" is also often used as a vague term, sometimes synonymous with "
395:
376:
360:
140:
54:
1190:
897:
757:
443:
was proposed as a maximally intelligent agent in this paradigm. However, AIXI is
391:
147:
88:
1798:
1750:
1508:
989:
913:
781:
749:
742:
603:
556:, ignoring the rest of the percept history. The agent function is based on the
436:
function" that influences how many descendants each agent is allowed to leave.
348:
328:
233:
223:
173:
109:
102:
78:
66:
31:
1762:
1133:
319:
vs "artificial" intelligence) and does not indicate that such a machine has a
1921:
1482:
1375:
1358:
380:
364:
324:
1429:
653:
624:
1828:
1590:
994:
978:
774:
707:
588:
575:
444:
1837:(2nd ed.). Upper Saddle River, New Jersey: Prentice Hall. Chapter 2.
1118:
1806:
1694:
1331:"Why AlphaZero's Artificial Intelligence Has Trouble With the Real World"
933:
929:
903:
510:
411:
386:
Goals can be explicitly defined or induced. If the AI is programmed for "
312:
70:
1871:
10.1002/(SICI)1098-111X(199806)13:6<453::AID-INT1>3.0.CO;2-K
1581:
1258:"Generative adversarial networks: What GANs are and how they've evolved"
908:
428:
177:
62:
50:
1632:"Competing For The Future With Intelligent Agents... And A Confession"
1359:"Mapping the Landscape of Human-Level Artificial General Intelligence"
544:
516:
The program agent, instead, maps every possible percept to an action.
981:– making data on the Web available for automated processing by agents
432:
372:
344:
123:
Intelligent agents in artificial intelligence are closely related to
98:
283:. Statements consisting only of original research should be removed.
209:
It also defines the field of "artificial intelligence research" as:
1033:
1031:
702:
In 2013, Alexander
Wissner-Gross published a theory pertaining to
1057:
Bringsjord, Selmer; Govindarajulu, Naveen Sundar (12 July 2018).
703:
506:
168:) to distinguish them from their real-world implementations. An
131:, and versions of the intelligent agent paradigm are studied in
53:
in order to achieve goals, and may improve its performance with
1832:
1285:"Artificial Intelligence Will Do What We Ask. That's a Problem"
1105:
Bull, Larry (1999). "On model-based evolutionary computation".
1078:"Artificial Intelligence Will Do What We Ask. That's a Problem"
1028:
804:
to represent short- and long-term memory, age, forgetting, etc.
219:
1683:
SAE International Journal of Connected and Automated Vehicles
1229:
1217:
1063:
The Stanford Encyclopedia of Philosophy (Summer 2020 Edition)
860:
183:
180:
computer program that carries out tasks on behalf of users).
136:
82:
1710:"Inside Waymo's Secret World for Training Self-Driving Cars"
780:
Learn and improve through interaction with the environment (
1205:
752:, IA systems should exhibit the following characteristics:
440:
320:
74:
1749:
Yang, Guoqing; Wu, Zhaohui; Li, Xiumei; Chen, Wei (2003).
1676:
552:
Simple reflex agents act only on the basis of the current
455:
A simple agent program can be defined mathematically as a
69:
is considered an example of an intelligent agent, as is a
1736:
In International Conference on Cyber Warfare and Security
1677:
Hallerbach, S.; Xia, Y.; Eberle, U.; Koester, F. (2018).
587:
A model-based reflex agent should maintain some sort of
73:, as is any system that meets the definition, such as a
1056:
1392:"Eye-catching advances in some AI fields are not real"
528:
1751:"SVE: embedded agent based smart vehicle environment"
1738:. Academic Conferences International Limited: 594-XI.
1733:
469:
1854:"Introduction: Hybrid intelligent adaptive systems"
1658:
732:
509:of individual options, deduction over logic rules,
222:way. Optional desiderata include that the agent be
61:. An intelligent agent may be simple or complex: A
1544:
1100:
1098:
494:
1545:Wissner-Gross, A. D.; Freer, C. E. (2013-04-19).
1355:
1919:
523:
120:'s behavior is shaped by a "fitness function".
37:In intelligence and artificial intelligence, an
1095:
1434:. Repin: Bruckner Publishing. pp. 42–59.
1176:
427:rewards for incremental progress in learning.
1823:
1495:
1415:
1211:
1156:
1139:
1069:
1037:
832:to certain ideas, incidents, or controversies
533:
1858:International Journal of Intelligent Systems
1748:
1509:https://doi.org/10.1016/j.artint.2018.01.002
1457:"Artificial intelligence: A modern approach"
662:responsible for selecting external actions.
570:
1421:
883:Artificial intelligence systems integration
1891:(2nd ed.). Cambridge, MA: MIT Press.
1834:Artificial Intelligence: A Modern Approach
1603:
842:this issue before removing this message.
343:(which is defined in terms of "goals") or
190:Artificial Intelligence: A Modern Approach
184:As a definition of artificial intelligence
112:of the objective function. For example, a
1869:
1580:
1570:
1472:
1374:
1282:
1075:
672:
299:Learn how and when to remove this message
213:"The study and design of rational agents"
1797:
1235:
1223:
777:in terms of behavior, error and success.
713:
652:
623:
602:
574:
543:
495:{\displaystyle f:P^{\ast }\rightarrow A}
331:(i.e., it does not imply John Searle's "
87:
19:For the term in intelligent design, see
1851:
1664:
1454:
1427:
1052:
1050:
1048:
1046:
619:
539:
1920:
1629:
1522:"A Universal Formula for Intelligence"
1389:
1152:
1150:
1148:
1076:Wolchover, Natalie (30 January 2020).
900:– a practical field for implementation
347:(which uses the same definition of a "
16:Software agent which acts autonomously
1886:
677:
354:
1707:
1520:Box, Geeks out of the (2019-12-04).
1104:
1043:
1021:
812:
787:Learn quickly from large amounts of
598:
267:Relevant discussion may be found on
244:
1791:
1757:. Vol. 2. pp. 1745–1749.
1519:
1283:Wolchover, Natalie (January 2020).
1145:
836:create a more balanced presentation
529:Russell and Norvig's classification
226:, and that the agent be capable of
13:
1142:, pp. 4–5, 32, 35, 36 and 56.
648:
201:It defines a "rational agent" as:
14:
1939:
1906:
450:
925:Friendly artificial intelligence
817:
733:Alternative definitions and uses
725:display a form of intelligence.
680:defines four classes of agents:
628:Model-based, utility-based agent
249:
1742:
1727:
1701:
1670:
1649:
1623:
1604:Poole, David; Mackworth, Alan.
1597:
1538:
1513:
1501:
1489:
1455:Nilsson, Nils J. (April 1996).
1448:
1409:
1390:Hutson, Matthew (27 May 2020).
1383:
1349:
1323:
1311:
1302:
1276:
1250:
1241:
939:GOAL agent programming language
932:– IA implemented with adaptive
808:
560:: "if condition, then action".
404:generative adversarial networks
1572:10.1103/PhysRevLett.110.168702
1170:
1161:
1008:
486:
439:The mathematical formalism of
1:
1708:Madrigal, Story by Alexis C.
970:– multiple interactive agents
607:Model-based, goal-based agent
524:Classes of intelligent agents
379:", "objective function", or "
240:
1689:(2). SAE International: 93.
1474:10.1016/0004-3702(96)00007-0
1431:Design of Agent-Based Models
1191:10.1016/j.bushor.2018.08.004
1061:. In Edward N. Zalta (ed.).
170:autonomous intelligent agent
7:
1157:Russell & Norvig (2003)
870:
793:Have memory-based exemplar
534:Russell & Norvig (2003)
279:the claims made and adding
162:abstract intelligent agents
92:Simple reflex agent diagram
10:
1944:
739:virtual personal assistant
717:
358:
25:
18:
1763:10.1109/ITSC.2003.1252782
1496:Russell & Norvig 2003
1416:Russell & Norvig 2003
1212:Russell & Norvig 2003
1140:Russell & Norvig 2003
1059:"Artificial Intelligence"
1038:Russell & Norvig 2003
944:Hybrid intelligent system
571:Model-based reflex agents
341:mathematical optimization
47:perceives its environment
1547:"Causal Entropic Forces"
1376:10.1609/aimag.v33i1.2322
1001:
797:and retrieval capacities
710:for intelligent agents.
697:
657:A general learning agent
579:Model-based reflex agent
139:, and the philosophy of
26:Not to be confused with
1928:Artificial intelligence
1551:Physical Review Letters
1461:Artificial Intelligence
1428:Salamon, Tomas (2011).
959:JACK Intelligent Agents
893:Cognitive architectures
228:belief-desire-intention
193:defines an "agent" as
28:Artificial intelligence
1801:(September 22, 2015).
1630:Fingar, Peter (2018).
974:Reinforcement learning
673:Weiss's classification
658:
629:
608:
580:
549:
496:
424:Reinforcement learning
420:evolutionary computing
388:reinforcement learning
269:Talk:Intelligent agent
215:
207:
199:
118:evolutionary algorithm
114:reinforcement learning
93:
1119:10.1007/s005000050055
968:multiple-agent system
714:Hierarchies of agents
656:
627:
606:
584:"model-based agent".
578:
558:condition-action rule
547:
497:
359:Further information:
211:
203:
195:
143:, as well as in many
91:
1852:Kasabov, N. (1998).
1695:10.4271/2018-01-1066
1526:Geeks out of the box
878:Ambient intelligence
773:Are able to analyze
620:Utility-based agents
540:Simple reflex agents
467:
363:(economics) and
333:strong AI hypothesis
21:Intelligent designer
1563:2013PhRvL.110p8702W
949:Intelligent control
760:rules incrementally
548:Simple reflex agent
45:) is an agent that
1889:Multiagent systems
1887:Weiss, G. (2013).
1825:Russell, Stuart J.
1264:. 26 December 2019
964:Multi-agent system
954:Intelligent system
720:Multi-agent system
659:
630:
609:
581:
550:
492:
355:Objective function
260:possibly contains
155:social simulations
94:
1898:978-0-262-01889-0
1441:978-80-904661-1-1
1179:Business Horizons
1022:Inline references
985:Social simulation
865:self-driving cars
856:
855:
834:. Please help to
826:This section may
599:Goal-based agents
309:
308:
301:
262:original research
145:interdisciplinary
133:cognitive science
39:intelligent agent
1935:
1902:
1883:
1873:
1848:
1820:
1792:Other references
1785:
1784:
1746:
1740:
1739:
1731:
1725:
1724:
1722:
1720:
1705:
1699:
1698:
1674:
1668:
1662:
1656:
1653:
1647:
1646:
1644:
1642:
1627:
1621:
1620:
1618:
1616:
1601:
1595:
1594:
1584:
1574:
1542:
1536:
1535:
1533:
1532:
1517:
1511:
1505:
1499:
1498:, pp. 46–54
1493:
1487:
1486:
1476:
1467:(1–2): 369–380.
1452:
1446:
1445:
1425:
1419:
1413:
1407:
1406:
1404:
1402:
1387:
1381:
1380:
1378:
1353:
1347:
1346:
1344:
1342:
1327:
1321:
1315:
1309:
1306:
1300:
1299:
1297:
1295:
1280:
1274:
1273:
1271:
1269:
1254:
1248:
1245:
1239:
1233:
1227:
1221:
1215:
1209:
1203:
1202:
1174:
1168:
1165:
1159:
1154:
1143:
1137:
1131:
1130:
1102:
1093:
1092:
1090:
1088:
1073:
1067:
1066:
1054:
1041:
1035:
1015:
1012:
919:Federated search
888:Autonomous agent
851:
848:
821:
820:
813:
756:Accommodate new
635:utility function
501:
499:
498:
493:
485:
484:
396:fitness function
377:utility function
361:utility function
304:
297:
293:
290:
284:
281:inline citations
253:
252:
245:
141:practical reason
49:, takes actions
1943:
1942:
1938:
1937:
1936:
1934:
1933:
1932:
1918:
1917:
1909:
1899:
1845:
1817:
1799:Domingos, Pedro
1794:
1789:
1788:
1773:
1747:
1743:
1732:
1728:
1718:
1716:
1706:
1702:
1675:
1671:
1663:
1659:
1654:
1650:
1640:
1638:
1628:
1624:
1614:
1612:
1602:
1598:
1543:
1539:
1530:
1528:
1518:
1514:
1506:
1502:
1494:
1490:
1453:
1449:
1442:
1426:
1422:
1414:
1410:
1400:
1398:
1388:
1384:
1354:
1350:
1340:
1338:
1335:Quanta Magazine
1329:
1328:
1324:
1316:
1312:
1307:
1303:
1293:
1291:
1289:Quanta Magazine
1281:
1277:
1267:
1265:
1256:
1255:
1251:
1246:
1242:
1234:
1230:
1222:
1218:
1210:
1206:
1175:
1171:
1166:
1162:
1155:
1146:
1138:
1134:
1103:
1096:
1086:
1084:
1082:Quanta Magazine
1074:
1070:
1055:
1044:
1036:
1029:
1024:
1019:
1018:
1013:
1009:
1004:
999:
898:Cognitive radio
873:
852:
846:
843:
822:
818:
811:
758:problem solving
743:software agents
735:
722:
716:
700:
675:
651:
649:Learning agents
622:
601:
573:
542:
531:
526:
480:
476:
468:
465:
464:
453:
416:neural networks
392:reward function
368:
357:
305:
294:
288:
285:
266:
254:
250:
243:
186:
174:software agents
148:socio-cognitive
35:
24:
17:
12:
11:
5:
1941:
1931:
1930:
1916:
1915:
1908:
1907:External links
1905:
1904:
1903:
1897:
1884:
1864:(6): 453–454.
1849:
1843:
1821:
1816:978-0465065707
1815:
1793:
1790:
1787:
1786:
1771:
1741:
1726:
1700:
1669:
1657:
1648:
1622:
1596:
1557:(16): 168702.
1537:
1512:
1500:
1488:
1447:
1440:
1420:
1408:
1396:Science | AAAS
1382:
1348:
1322:
1310:
1301:
1275:
1249:
1240:
1228:
1216:
1204:
1169:
1160:
1144:
1132:
1107:Soft Computing
1094:
1068:
1042:
1026:
1025:
1023:
1020:
1017:
1016:
1006:
1005:
1003:
1000:
998:
997:
992:
990:Software agent
987:
982:
976:
971:
961:
956:
951:
946:
941:
936:
927:
922:
916:
914:Embodied agent
911:
906:
901:
895:
890:
885:
880:
874:
872:
869:
854:
853:
847:September 2023
838:. Discuss and
825:
823:
816:
810:
807:
806:
805:
798:
791:
785:
778:
771:
761:
750:Nikola Kasabov
734:
731:
718:Main article:
715:
712:
699:
696:
695:
694:
691:
688:
685:
674:
671:
650:
647:
621:
618:
600:
597:
589:internal model
572:
569:
541:
538:
530:
527:
525:
522:
503:
502:
491:
488:
483:
479:
475:
472:
452:
451:Agent function
449:
356:
353:
349:rational agent
307:
306:
257:
255:
248:
242:
239:
185:
182:
110:expected value
103:rational agent
67:control system
32:Embodied agent
15:
9:
6:
4:
3:
2:
1940:
1929:
1926:
1925:
1923:
1914:
1911:
1910:
1900:
1894:
1890:
1885:
1881:
1877:
1872:
1867:
1863:
1859:
1855:
1850:
1846:
1844:0-13-790395-2
1840:
1836:
1835:
1830:
1829:Norvig, Peter
1826:
1822:
1818:
1812:
1808:
1804:
1800:
1796:
1795:
1782:
1778:
1774:
1772:0-7803-8125-4
1768:
1764:
1760:
1756:
1752:
1745:
1737:
1730:
1715:
1711:
1704:
1696:
1692:
1688:
1684:
1680:
1673:
1666:
1661:
1652:
1637:
1633:
1626:
1611:
1607:
1600:
1592:
1588:
1583:
1578:
1573:
1568:
1564:
1560:
1556:
1552:
1548:
1541:
1527:
1523:
1516:
1510:
1504:
1497:
1492:
1484:
1480:
1475:
1470:
1466:
1462:
1458:
1451:
1443:
1437:
1433:
1432:
1424:
1417:
1412:
1397:
1393:
1386:
1377:
1372:
1368:
1364:
1360:
1352:
1336:
1332:
1326:
1319:
1314:
1305:
1290:
1286:
1279:
1263:
1259:
1253:
1244:
1237:
1236:Domingos 2015
1232:
1225:
1224:Domingos 2015
1220:
1214:, p. 27.
1213:
1208:
1200:
1196:
1192:
1188:
1184:
1180:
1173:
1164:
1158:
1153:
1151:
1149:
1141:
1136:
1128:
1124:
1120:
1116:
1112:
1108:
1101:
1099:
1083:
1079:
1072:
1064:
1060:
1053:
1051:
1049:
1047:
1039:
1034:
1032:
1027:
1011:
1007:
996:
993:
991:
988:
986:
983:
980:
977:
975:
972:
969:
965:
962:
960:
957:
955:
952:
950:
947:
945:
942:
940:
937:
935:
931:
928:
926:
923:
920:
917:
915:
912:
910:
907:
905:
902:
899:
896:
894:
891:
889:
886:
884:
881:
879:
876:
875:
868:
866:
862:
850:
841:
837:
833:
831:
824:
815:
814:
803:
799:
796:
792:
790:
786:
783:
779:
776:
772:
770:
766:
762:
759:
755:
754:
753:
751:
748:According to
746:
744:
740:
730:
726:
721:
711:
709:
705:
692:
689:
686:
683:
682:
681:
679:
670:
667:
663:
655:
646:
642:
640:
636:
626:
617:
615:
605:
596:
593:
590:
585:
577:
568:
565:
561:
559:
555:
546:
537:
535:
521:
517:
514:
512:
508:
489:
481:
477:
473:
470:
463:
462:
461:
458:
448:
446:
442:
437:
434:
430:
425:
421:
417:
413:
408:
405:
399:
397:
393:
390:", it has a "
389:
384:
382:
381:loss function
378:
374:
367:(mathematics)
366:
365:loss function
362:
352:
350:
346:
342:
336:
334:
330:
329:understanding
326:
325:consciousness
322:
318:
314:
303:
300:
292:
289:February 2023
282:
278:
274:
270:
264:
263:
258:This section
256:
247:
246:
238:
235:
231:
229:
225:
221:
214:
210:
206:
202:
198:
194:
192:
191:
181:
179:
175:
171:
167:
163:
158:
156:
153:and computer
152:
149:
146:
142:
138:
134:
130:
126:
121:
119:
115:
111:
106:
104:
100:
90:
86:
84:
80:
76:
72:
68:
64:
60:
57:or acquiring
56:
52:
48:
44:
40:
33:
29:
22:
1888:
1861:
1857:
1833:
1802:
1754:
1744:
1735:
1729:
1717:. Retrieved
1714:The Atlantic
1713:
1703:
1686:
1682:
1672:
1665:Kasabov 1998
1660:
1651:
1639:. Retrieved
1636:Forbes Sites
1635:
1625:
1613:. Retrieved
1609:
1599:
1582:1721.1/79750
1554:
1550:
1540:
1529:. Retrieved
1525:
1515:
1503:
1491:
1464:
1460:
1450:
1430:
1423:
1418:, p. 33
1411:
1399:. Retrieved
1395:
1385:
1366:
1362:
1351:
1339:. Retrieved
1334:
1325:
1313:
1304:
1292:. Retrieved
1288:
1278:
1266:. Retrieved
1261:
1252:
1243:
1238:, Chapter 7.
1231:
1226:, Chapter 5.
1219:
1207:
1185:(1): 15–25.
1182:
1178:
1172:
1163:
1135:
1113:(2): 76–82.
1110:
1106:
1085:. Retrieved
1081:
1071:
1062:
1010:
995:Software bot
979:Semantic Web
930:Fuzzy agents
857:
844:
830:undue weight
827:
809:Applications
747:
736:
727:
723:
708:Intelligence
701:
678:Weiss (2013)
676:
668:
664:
660:
643:
634:
631:
610:
594:
586:
582:
566:
562:
557:
551:
532:
518:
515:
504:
454:
445:uncomputable
438:
409:
400:
385:
369:
337:
310:
295:
286:
259:
232:
216:
212:
208:
204:
200:
196:
188:
187:
169:
165:
161:
159:
122:
107:
95:
51:autonomously
42:
38:
36:
1807:Basic Books
1615:28 November
1610:artint.info
1363:AI Magazine
1318:Martin Ford
1262:VentureBeat
934:fuzzy logic
904:Cybernetics
511:fuzzy logic
412:symbolic AI
317:"synthetic"
313:Turing test
71:human being
1531:2022-10-11
1040:, chpt. 2.
909:DAYDREAMER
802:parameters
782:embodiment
775:themselves
639:comparison
429:Yann LeCun
273:improve it
241:Advantages
230:analysis.
178:autonomous
63:thermostat
1880:120318478
1781:110177067
1719:14 August
1483:0004-3702
1369:(1): 25.
1199:158433736
769:real time
487:→
482:∗
433:AlphaZero
345:economics
277:verifying
271:. Please
129:economics
99:economics
65:or other
59:knowledge
1922:Category
1913:Coneural
1831:(2003).
1591:23679649
871:See also
614:planning
457:function
327:or true
224:rational
151:modeling
55:learning
30:(AI) or
1641:18 June
1559:Bibcode
1401:18 June
1341:18 June
1294:18 June
1268:18 June
1127:9699920
1087:21 June
840:resolve
795:storage
767:and in
704:Freedom
554:percept
513:, etc.
507:utility
418:and to
81:, or a
1895:
1878:
1841:
1813:
1779:
1769:
1589:
1481:
1438:
1337:. 2018
1197:
1125:
765:online
763:Adapt
410:While
234:Kaplan
220:robust
137:ethics
125:agents
1876:S2CID
1777:S2CID
1195:S2CID
1123:S2CID
1002:Notes
861:Waymo
828:lend
800:Have
698:Other
83:biome
79:state
1893:ISBN
1839:ISBN
1811:ISBN
1767:ISBN
1721:2020
1643:2020
1617:2018
1587:PMID
1479:ISSN
1436:ISBN
1403:2020
1343:2020
1296:2020
1270:2020
1089:2020
966:and
789:data
706:and
441:AIXI
383:".)
351:").
321:mind
176:(an
77:, a
75:firm
1866:doi
1759:doi
1691:doi
1577:hdl
1567:doi
1555:110
1469:doi
1371:doi
1187:doi
1115:doi
275:by
166:AIA
127:in
105:".
101:, "
1924::
1874:.
1862:13
1860:.
1856:.
1827:;
1809:.
1805:.
1775:.
1765:.
1753:.
1712:.
1685:.
1681:.
1634:.
1608:.
1585:.
1575:.
1565:.
1553:.
1549:.
1524:.
1477:.
1465:82
1463:.
1459:.
1394:.
1367:33
1365:.
1361:.
1333:.
1287:.
1260:.
1193:.
1183:62
1181:.
1147:^
1121:.
1109:.
1097:^
1080:.
1045:^
1030:^
422:.
373:Go
323:,
157:.
135:,
85:.
43:IA
1901:.
1882:.
1868::
1847:.
1819:.
1783:.
1761::
1723:.
1697:.
1693::
1687:1
1667:.
1645:.
1619:.
1593:.
1579::
1569::
1561::
1534:.
1485:.
1471::
1444:.
1405:.
1379:.
1373::
1345:.
1298:.
1272:.
1201:.
1189::
1129:.
1117::
1111:3
1091:.
1065:.
849:)
845:(
784:)
490:A
478:P
474::
471:f
302:)
296:(
291:)
287:(
265:.
164:(
41:(
34:.
23:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.