1 00:00:03,423 --> 00:00:07,914 As we shall see, general game playing is an interesting application in its own right. 2 00:00:08,366 --> 00:00:12,051 It's intellectually engaging and more than a little fun. 3 00:00:12,614 --> 00:00:14,166 But it's much more than that. 4 00:00:15,260 --> 00:00:23,867 It serves as an analog for applications of logic in other areas such as business and law, science and engineering, and other areas. 5 00:00:24,109 --> 00:00:33,493 More fundamentally, it raises questions about the nature of intelligence and serves as a laboratory which through evaluate competing approaches to intelligence. 6 00:00:36,270 --> 00:00:43,861 Game descriptions provide full information about a world and determine optimal strategies as a baseline for evaluating agent behavior. 7 00:00:44,547 --> 00:00:51,939 By its nature, the general game playing setting can be used to evaluate problem solving strategies and by extension theories of intelligence 8 00:00:52,229 --> 00:00:59,032 by taking into account representation, incompleteness of information and ressource bounds. 9 00:01:00,634 --> 00:01:08,816 The idea of using applications like general game playing for such purposes is not new. 10 00:01:09,267 --> 00:01:14,572 It was in 1958 that John McCarty invented the concept of the "advice taker". 11 00:01:15,100 --> 00:01:20,131 The idea was simple: he wanted a machine that he could program by description. 12 00:01:20,487 --> 00:01:28,843 He would describe the intended environment and the desired goal, and the machine would use that information in determining its behavior. 13 00:01:29,101 --> 00:01:31,605 There would be no programming in the traditional sense. 14 00:01:32,411 --> 00:01:38,072 McCarthy presented his concept in a paper that has become a classic in the field of AI. Here is a quote: 15 00:01:38,514 --> 00:01:56,652 16 00:01:58,362 --> 00:02:02,679 It was an ambitious goal, but those were times of high hopes and grand ambitions. 17 00:02:03,800 --> 00:02:17,070 The idea caught the imaginations of numerous subsequent researchers, notably Bob Kowalski, the high priest of logic programming, and Ed Feigenbaum, the inventor of knowledge engineering. 18 00:02:17,870 --> 00:02:24,509 In a paper written in 1974, Feigenbaum gave us the most forceful statement of McCarthy's ideal: 19 00:02:25,992 --> 00:03:03,022 20 00:03:05,266 --> 00:03:11,730 Okay so one final remark: some have argued that the way to achieve intelligent behavior is through specialization. 21 00:03:13,069 --> 00:03:17,763 That may work so long as the assumptions one makes in building such a assumptions are true. 22 00:03:18,516 --> 00:03:27,053 For a general intelligence however, general intellectual capabilities are needed and such systems should be capable of performing well in a wide variety of tasks. 23 00:03:28,221 --> 00:03:30,351 To paraphrase the words of Robert Heinlein: 24 00:03:30,726 --> 00:03:58,567 25 00:03:59,811 --> 00:04:05,360 Those of us who are more interested in artificial intelligence than artificial insects agree with Heinlein.