Navigation

Discovery without Objectives

Description: 
This lecture was given by Assoc. Prof. Ken Stanley, University of Central Florida, when he visited RMIT University's School of Computer Science and IT in August 2012.

This lecture was given by Assoc. Prof. Ken Stanley, University of Central Florida, when he visited RMIT University's School of Computer Science and IT in August 2012. The talk was sponsored by NICTA, ACM and RMIT University.

In this talk, titled 'Discovery without Objectives', Ken Stanley offers an alternative approach to innovation, creativity, and discovery, by contemplating the counterintuitive idea that setting an objective can often limit or even block its own achievement. Recent experiments with evolutionary algorithms suggest that search (a metaphor for innovation and discovery in general) is sometimes most effective when it is not explicitly seeking an objective. In particular, through several experiments in interactive evolution, and with an algorithm called 'novelty search', a picture of innovation is emerging in which defined objectives can help to guide us one step away from our present understanding, yet ultimately become handcuffs that also blind us to essential orthogonal discoveries on the road to long-term innovation. While the implications of these insights for reaching our highest goals are in part sobering, the silver lining is that much can be gained by liberating ourselves from the temptation to frame all our projects in terms of what they ultimately aim to achieve. Instead, with evidence in hand, we can exploit the structure of the unknown by orienting ourselves towards discovery and away from the shackles of mandated outcomes.

This talk was part of the ACM's Distinguished Lecture Series and was also designated a NICTA Big Picture Seminar.

iTunes Category: 
Computer Science
Author: 
RMIT University
Keywords: 
Artificial Intelligence, computational intelligence, evolutionary algorithms, artificial neural networks, evolving neural networks,