Where are Bottlenecks in NK Fitness Landscapes?

TL;DR

This paper introduces the Fitness Cloud (FC), a new visualization tool for analyzing fitness landscapes, particularly identifying bottlenecks that hinder local search algorithms like hill-climbing and simulated annealing.

Contribution

The paper presents the Fitness Cloud as a novel generalization of fitness correlation, enabling visualization of bottlenecks and neutrality in fitness landscapes.

Findings

Identified bottlenecks in NK fitness landscapes using FC.

Demonstrated how neutrality information from FC can improve local search heuristics.

Validated FC's effectiveness on well-known fitness landscapes.

Abstract

Usually the offspring-parent fitness correlation is used to visualize and analyze some caracteristics of fitness landscapes such as evolvability. In this paper, we introduce a more general representation of this correlation, the Fitness Cloud (FC). We use the bottleneck metaphor to emphasise fitness levels in landscape that cause local search process to slow down. For a local search heuristic such as hill-climbing or simulated annealing, FC allows to visualize bottleneck and neutrality of landscapes. To confirm the relevance of the FC representation we show where the bottlenecks are in the well-know NK fitness landscape and also how to use neutrality information from the FC to combine some neutral operator with local search heuristic.