Human-assisted computation also called humanized or human-based computing harnesses human intelligence to solve computational problems that are beyond the scope of existing Artificial Intelligence algorithms by outsourcing certain steps to humans. This approach exploits differences in abilities between humans and computer agents creating a symbiotic human-computer interaction. Human assistance can be helpful in solving many AI-complete problems, which by definition are tasks which are impossible for computers but possible for humans. For computationally difficult tasks such as image recognition, human-based computation plays a central role in optimizing Deep Learning-based Artificial Intelligence systems, as can be seen in many citizen-science projects such as Zooniverse.
This book focuses on well-known and new methodologies of optimization techniques in human-assisted computing that are used to resolve some of the very complicated and hard problems we face today. Excitingly techniques originally developed for solutions in engineering, science and technology are being applied to areas of economics, finance and, especially the social sciences. The book reviews present and developing deep model-based methods of mathematics and the less model-based, also so-called smart or intelligent algorithms, with their roots in engineering, computer science or informatics and how combing the strengths of humans and computers can create powerful solutions. Often called heuristics and model-free, they are less rigorous mathematically, enabling the integration of bio-inspired approaches to efficiently cope with AI-complete problems. The rise of these algorithms from Artificial Intelligence research happened in parallel with the powerful progress in model-based mathematics. Often called Statistical Learning, Machine Learning, Metaheuristics or Operational Research, model-free and model-based traditions gain from the synergistic effects of humanized computing, revealing solutions to urgent real-life challenges across numerous fields. Bridging several scientific disciplines, this book will be essential reading for scientists, engineers and applied mathematicians alike.
This book focuses on well-known and new methodologies of optimization techniques in human-assisted computing that are used to resolve some of the very complicated and hard problems we face today. Excitingly techniques originally developed for solutions in engineering, science and technology are being applied to areas of economics, finance and, especially the social sciences. The book reviews present and developing deep model-based methods of mathematics and the less model-based, also so-called smart or intelligent algorithms, with their roots in engineering, computer science or informatics and how combing the strengths of humans and computers can create powerful solutions. Often called heuristics and model-free, they are less rigorous mathematically, enabling the integration of bio-inspired approaches to efficiently cope with AI-complete problems. The rise of these algorithms from Artificial Intelligence research happened in parallel with the powerful progress in model-based mathematics. Often called Statistical Learning, Machine Learning, Metaheuristics or Operational Research, model-free and model-based traditions gain from the synergistic effects of humanized computing, revealing solutions to urgent real-life challenges across numerous fields. Bridging several scientific disciplines, this book will be essential reading for scientists, engineers and applied mathematicians alike.
Dieser Download kann aus rechtlichen Gründen nur mit Rechnungsadresse in A, D ausgeliefert werden.