Comparing Problem Solving Strategies for NP-hard Optimization Problems

Evaluating genetic algorithms through the approximability hierarchy

“…Some authors have already combined genetic algorithms and the approximability hierarchy, but only in the context of justifying the difficulty of dealing with a given problem. Good examples are [42,43,44,45,46,20,47], where the authors first prove the approximability class a problem belongs to, and then they provide a genetic algorithm to solve it. However, there has been no previous attempt to analyze the effectiveness of genetic algorithms depending on the approximation class.…”

Analysis and Augmentation of Human Performance on Telerobotic Search Problems

“…Human performance on NP-hard problems is an interdisciplinary research area that combines computer science, psychology and neuroscience [1]. Humans can solve many forms of complex problems efficiently; therefore, researchers have been investigating human strategies for determining more efficient algorithms, understanding the underlying cognitive processes, and more specifically, how human cognition deals with NP-hard problems (e.g., traveling salesman problem (TSP) , Vertex Cover, and Knapsack) [1], [2].…”

“…The sensing coverage (the red triangle) of the robot is defined as f C (X R ), the percentage of area covered by the sensor. 1 Definition 2 (Subgoal Coverage): The robot discrete position (X k = {x k , y k , θ k }) and the measurement data (z k ) are recorded during the search task, 2 where k ∈ S = {1, .., K }. The subgoals index and subgoal positions are denoted as I and X I , respectively.…”

“…The details of computing coverage from sensing data and a grid map can be found in[5] 2. The robot's continuous trajectory was sampled via optical encoders, while the measurement was sampled via a Kinect sensor at 4HZ.3 C/T means the coverage (C) divided by the traveling time (T).VOLUME 8, 2020…”

A Survey on Nature-Inspired Medical Image Analysis: A Step Further in Biomedical Data Integration

“…Indeed, fascinating processes occur in nature, giving rise to an ever-increasing scientific interest [1]. In real-world applications, many problems are complex or computationally intensive for traditional exhaustive algorithmic methods [2], such as in the case of NP-hard problems [3]. Relying on natural mechanisms, information processing in nature is generally performed by means of distributed and self-organised approaches without requiring a centralised control [4].…”