We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
To save content items to your account,
please confirm that you agree to abide by our usage policies.
If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account.
Find out more about saving content to .
To save content items to your Kindle, first ensure no-reply@cambridge.org
is added to your Approved Personal Document E-mail List under your Personal Document Settings
on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part
of your Kindle email address below.
Find out more about saving to your Kindle.
Note you can select to save to either the @free.kindle.com or @kindle.com variations.
‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi.
‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.
Convolutional sequence to sequence (CNN seq2seq) models have met success in abstractive summarization. However, their outputs often contain repetitive word sequences and logical inconsistencies, limiting the practicality of their application. In this paper, we find the reasons behind the repetition problem in CNN-based abstractive summarization through observing the attention map between the summaries with repetition and their corresponding source documents and mitigate the repetition problem. We propose to reduce the repetition in summaries by attention filter mechanism (ATTF) and sentence-level backtracking decoder (SBD), which dynamically redistributes attention over the input sequence as the output sentences are generated. The ATTF can record previously attended locations in the source document directly and prevent the decoder from attending to these locations. The SBD prevents the decoder from generating similar sentences more than once via backtracking at test. The proposed model outperforms the baselines in terms of ROUGE score, repeatedness, and readability. The results show that this approach generates high-quality summaries with minimal repetition and makes the reading experience better.
Event structures are central in Linguistics and Artificial Intelligence research: people can easily refer to changes in the world, identify their participants, distinguish relevant information, and have expectations of what can happen next. Part of this process is based on mechanisms similar to narratives, which are at the heart of information sharing. But it remains difficult to automatically detect events or automatically construct stories from such event representations. This book explores how to handle today's massive news streams and provides multidimensional, multimodal, and distributed approaches, like automated deep learning, to capture events and narrative structures involved in a 'story'. This overview of the current state-of-the-art on event extraction, temporal and casual relations, and storyline extraction aims to establish a new multidisciplinary research community with a common terminology and research agenda. Graduate students and researchers in natural language processing, computational linguistics, and media studies will benefit from this book.
Chapter 3 presents models in which artificial agents interact through games. It first introduces a typical mainstream economics model known as the prisoner’s dilemma, in which agents interact through a classic game theory game, and then contrast it with an artificial evolutionary game based on the same dilemma. In this evolutionary game, the dynamic evolution of a population of boundedly rational agents is represented and simulated using a genetic algorithm. It finally contrasts the assumptions of artificial economics against those of mainstream economics when modeling games.
There are various paradigms and programming methods suitable for different types of applications. The programming and simulation of an artificial economy can be carried out with standard methods, or through object-oriented programming, which is a very structured way of constructing computational objects, which are encapsulated in themselves, but can interact with other computational objects.
Chapter 8 presents the main positions in economics and in the social sciences regarding the agent/structure problem, and explores some contributions that can be made from artificial economics. First, it presents and discusses the individualist/reductionist, structuralist/holistic, and intermediate positions, regarding the agent/structure problem. Then presents simple artificial economics examples of the generation of endogenous preferences, agents' behavioral changes derived from their economic interaction, and of the demographic effects of the introduction of a market institution into an artificial economy.
Chapter 5 presents the subject matter of artificial intelligence, focusing on machine learning, where these machines are artificial agents. It presents simple examples of unsupervised learning, supervised learning, and reinforcement learning, and introduces notions of dynamic programming, Q-learning, and stochastic control. After that, it explores some links that can be established between artificial intelligence and the philosophy of mind, presenting and discussing the Turing test, the philosophical approaches of eliminativism and functionalism, and the problem of tacit knowledge.
Chapter 2 focuses on market interactions between artificial agents. First, it introduces two versions of the famous Sugarscape model, which is an excellent model for getting into artificial economics. In the first version, artificial agents are very simple and dedicated, in an environment in which there is only one resource called sugar, to collect and consume that resource to meet their metabolic needs. In the second version, the environment provides two resources, sugar and spice. And the artificial agents are more sophisticated, as they are not only dedicated to collecting and consuming such resources but also engage in market exchanges of them. Finally, it presents a static and a dynamic general equilibrium model of market economies, typical of mainstream economics, in a way that illustrates by contrast the assumptions of artificial economics versus the ones of mainstream economics when modeling markets.
Chapter 6 discusses the topic of artificial evolution. First, it introduces the concept of extrinsic adaptation and illustrates it through simple examples of genetic algorithms. It then introduces the concept of intrinsic adaptation and illustrates it through examples of evolution in artificial markets. Finally, introduces some basic concepts of the science of evolution related to genetic dynamic and natural selection, and presents and discusses the scope and limitations of the uses of evolutionary theory concepts and method in economics.
Chapter 4 focuses in the methodological and instrumental contrasts between artificial economics and mainstream economics. It discusses the mathematical methods of mainstream economics (centered around the use of optimization methods and systems of equations representations) versus the computational methods of artificial economics (characterized by the use of algorithms, software, and computer hardware). Presents basic notions on algorithms, recursion, and Turing machines. And discusses the methodological and instrumental differences between artificial economics and mainstream economics as derived from differences between classical mathematics and constructive mathematics.
Chapter 1 looks at how historically the concept of economic agent developed within mainstream economics, and how the concept of artificial agent emerged while cognitive science became the successor of behaviorism. Then, considering that artificial economics tries to build realistic models of artificial agents, it introduces the main models of mental architectures that derive from cognitive science, and some recent advances in neuroscience (specially within neuroeconomics, social neuroscience, and neurosociology) that relate directly to the economic and social behavior of individuals. Finally, it reviews some models and approaches that try to capture the cognitive, neurological, emotional, and social aspects of agents in an integrated way.
Chapter 7 introduces the subject matter of artificial complexity. First, it presents examples of artificial complexity by means of cellular automata. It presents one-dimensional cellular automata following Wolfram’s rules, and a two-dimensional cellular automaton in the form of a spatial evolutionary game.Then it introduces the concepts of complexity and emergence, as used in the science of complexity, and discusses some issues related to their definition and measurement. Finally, it discusses the scope and controversies around the application of the concepts and models of the science of complexity in economics.