错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Evolution of Artificial Intelligence, Continued

  • Lee Spector,
  • Ryan Bahlous-Boldi,
  • Paul Vander Vort,
  • Niko Lorantos

摘要

This chapter revisits issues raised in “Evolution of artificial intelligenceartificial intelligence,” which was published in the journal Artificial Intelligence in 2006. Two decades later, we consider ways in which evolutionary concepts and mechanisms have influenced the development of AI in the intervening years, and we discuss ways in which we might expect them to influence the field going forward. We focus on two strengths of evolutionary methods, their capacity for structural innovation and their ability to leverage adaptive diversity. We present examples of these strengths from our recent work and argue that they are indicative of contributions that evolutionary methods can be expected to make to the development of AI technologies in the future.