Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
24,352
result(s) for
"Prediction theory."
Sort by:
The unpredictability of gameplay
The Unpredictability of Gameplay explores the many forms of unpredictability in games and proposes a comprehensive theoretical framework for understanding and categorizing non-deterministic game mechanics. Rather than viewing all game mechanics with unpredictable outcomes as a single concept, Mark R. Johnson develops a three-part typology for such mechanics, distinguishing between randomness, chance, and luck in gameplay, assessing games that range from grand strategy and MMORPGs to slot machines and card games. He also explores forms of unanticipated unpredictability, where elements of games fail to function as intended and create new forms of gameplay in the process. Covering a range of game concepts using these frameworks, The Unpredictability of Gameplay then explores three illustrative case studies: 1) procedural generation, 2) replay value and grinding, and 3) player-made practices designed to reduce the level of luck in non-deterministic games. Throughout, Johnson demonstrates the importance of looking more deeply at unpredictability in games and game design and the various ways in which unpredictability manifests while offering an invaluable tool for game scholars and game designers seeking to integrate unpredictability into their work.
Book
A Review on PMsub.2.5 Sources, Mass Prediction, and Association Analysis: Research Opportunities and Challenges
Air pollution has long been one of the most life-threatening issues which has led to massive amounts of premature human death due to fatal diseases and environmental disasters. Several Sustainable Development Goals (SDGs) set up by the United Nations coincide with the solutions for air pollution reduction. To reach a sustainable future, researchers have conducted many theoretical analyses or case studies of air pollution at different places on the globe and proposed prudent strategies for obtaining an equilibrium between socioeconomic development and air pollution reduction. This research selected a substantial number of articles and existing review papers published between 2013 and 2024 and organized these publications into subfields. This research was focused on filling the gap between existing reviews and the state-of-the-art technologies in the last decade. To be informative and contextual, this review presented a pathway for readers to comprehend the research in three contiguous phases of air pollution analysis, from compositional apportionment and mass prediction of pollution to disclosing associations between pollution concentration and natural or anthropogenic factors. At the end of this review, the author highlighted several research fields which have been overlooked in previous reviews but will be increasingly important in the future.
Journal Article
Suboptimal foraging theory
Optimal foraging theory (OFT) is based on the ecological concept that organisms select behaviors that convey future fitness, and on the mathematical concept of optimization: finding the alternative that provides the best value of a fitness measure. As implemented in, for example, state-based dynamic modeling, OFT is powerful for one key problem of modern ecology: modeling behavior as a tradeoff among competing fitness elements such as growth, risk avoidance, and reproductive output. However, OFT is not useful for other modern problems such as representing feedbacks within systems of interacting, unique individuals: When we need to model foraging by each of many individuals that interact competitively or synergistically, optimization is impractical or impossible—there are no optimal behaviors. For such problems we can, however, still use the concept of future fitness to model behavior by replacing optimization with less precise (but perhaps more realistic) techniques for ranking alternatives. Instead of simplifying the systems we model until we can find optimal behavior, we can use theory based on inaccurate predictions, coarse approximations, and updating to produce good behavior in more complex and realistic contexts. This so-called state- and prediction-based theory (SPT) can, for example, produce realistic foraging decisions by each of many unique, interacting individuals when growth rates and predation risks vary over space and time. Because SPT lets us address more natural complexity and more realistic problems, it is more easily tested against more kinds of observation and more useful in management ecology. A simple foraging model illustrates how SPT readily accommodates complexities that make optimization intractable. Other models use SPT to represent contingent decisions (whether to feed or hide, in what patch) that are tradeoffs between growth and predation risk, when both growth and risk vary among hundreds of patches, vary unpredictably over time, depend on characteristics of the individuals, are subject to feedbacks from competition, and change over the daily light cycle. Modern ecology demands theory for tradeoff behaviors in complex contexts that produce feedbacks; when optimization is infeasible, we should not be afraid to use approximate fitness-seeking methods instead.
Journal Article
The human test : how predictability, creativity, and the quantum mind will redefine life in the age of AI
\"As data harvesting continues its exponential growth and computing power increases beyond what we can even imagine, AI and Big Data will be able to predict what we want for breakfast, what shoes we will buy, which political party we will vote for, and who we will fancy in a bar. Indeed, these deep-learning machines will know us better than we know ourselves. The day awaits when academia and big business will be able to quantify just how predictable each of us are. If indeed we are predictable like machines, to what extent are we alive, and under what definition? In The Human Test, quantum physicist Ron Folman unites findings from cognitive science, quantum physics, philosophy, and technology to offer a prescient look into this startling new era of human existence. If we are indeed found to be predictable, it will change everything we thought we knew about human nature. The Human Test strives for a new paradigm, and this new paradigm is found in human predictability. Existing monitors of general brain activity measure only very rudimentary aspects of the human experience, or what may be called human life; could the advent of a disruptive new technology bring about a much more insightful measure of who we are? While The Human Test describes a new and profound near-future impact of AI, it is rooted in a topic of paramount importance to humans: the enigma of consciousness and how it defines human life-or, what differentiates humans from all other life forms. Ultimately, The Human Test addresses some of the most fundamental questions about our species, starting with the most crucial: do we have free will, or are we merely machines? Fortunately, there is an antidote to predictability: creativity, that wondrous state of being original and inherently not predictable. Ultimately, Folman argues that if we can understand predictability, we can learn how to distance ourselves from it, and transcend to a purer idea of what human life should look like\"-- Provided by publisher.
BOOK
Forecasting gold price with the XGBoost algorithm and SHAP interaction values
Financial institutions, investors, mining companies and related firms need an effective accurate forecasting model to examine gold price fluctuations in order to make correct decisions. This paper proposes an innovative approach to accurately forecast gold price movements and to interpret predictions. First, it compares six machine learning models. These models include two very recent methods: the eXtreme Gradient Boosting (XGBoost) and CatBoost. The empirical findings indicate the superiority of XGBoost over other advanced machine learning models. Second, it proposes Shapley additive explanations (SHAP) in order to help policy makers to interpret the predictions of complex machine learning models and to examine the importance of various features that affect gold prices. Our results illustrate that the utilization of XGBoost along with SHAP approach could provide a significant boost in increasing the gold price forecasting performance.
Journal Article
Everyday chaos : technology, complexity, and how we're thriving in a new world of possibility
Modern science, the Internet, big data, and AI are each saying the same thing to us: the world is -- and always has been -- far more complex and unpredictable than we've allowed ourselves to see. As a result we're undergoing a sea change in our understanding of how things happen, and in our deepest strategies for predicting, preparing for, and managing our lives and our businesses. For example, machine learning allows us to make better predictions (think the weather, stock performance, online clicks) but we know less about why those predictions are right--and we need to get used to that. And in fact, over the past twenty years we've been unintentionally developing strategies that avoid anticipating what will happen so we don't have to depend on unreliable revenue forecasts, assumptions about customer needs, and hypotheses about how a product will be used. By embracing these strategies, we're flourishing by creating yet more possibilities and yet more unpredictability. In wide-ranging stories and characteristically all-encompassing syntheses, technology researcher, internet expert, and philosopher David Weinberger reveals the trends that hide in so many aspects of our lives--and shows us how they matter.-- Provided by publisher
Book
Point and interval prediction of crude oil futures prices based on chaos theory and multiobjective slime mold algorithm
Crude oil is the most important energy source in the world, and fluctuations in oil prices can significantly influence investors, companies, and governments. However, crude oil prices have numerous characteristics, including randomness, sudden structural changes, intrinsic nonlinearity, volatility, and chaotic nature. This makes the accurate forecasting of crude oil prices a difficult and challenging task. In this paper, a hybrid prediction model for crude oil futures prices is proposed, the accuracy and robustness of which are demonstrated via controlled experiments and sensitivity analysis. This study uses a new data denoising method for data processing to improve the accuracy and stability of the predictions of crude oil prices. Furthermore, the chaotic time-series prediction method, shallow neural networks, linear model prediction methods, and deep learning methods are adopted as submodels. The results of interval forecasts with narrow widths and high prediction accuracies are derived by introducing a confidence interval adjustment coefficient. The results of the simulation experiments indicate that the proposed hybrid prediction model exhibits higher accuracy and efficiency, as well as better robustness of the forecasting than the control models. In summary, the proposed forecasting framework can derive accurate point and interval forecasts and provide a valuable reference for the price forecasting of crude oil futures.
Journal Article
Statistical and Machine Learning forecasting methods: Concerns and ways forward
Machine Learning (ML) methods have been proposed in the academic literature as alternatives to statistical ones for time series forecasting. Yet, scant evidence is available about their relative performance in terms of accuracy and computational requirements. The purpose of this paper is to evaluate such performance across multiple forecasting horizons using a large subset of 1045 monthly time series used in the M3 Competition. After comparing the post-sample accuracy of popular ML methods with that of eight traditional statistical ones, we found that the former are dominated across both accuracy measures used and for all forecasting horizons examined. Moreover, we observed that their computational requirements are considerably greater than those of statistical methods. The paper discusses the results, explains why the accuracy of ML models is below that of statistical ones and proposes some possible ways forward. The empirical results found in our research stress the need for objective and unbiased ways to test the performance of forecasting methods that can be achieved through sizable and open competitions allowing meaningful comparisons and definite conclusions.
Journal Article
Struct2Graph: a graph attention network for structure based predictions of protein–protein interactions
Background
Development of new methods for analysis of protein–protein interactions (PPIs) at molecular and nanometer scales gives insights into intracellular signaling pathways and will improve understanding of protein functions, as well as other nanoscale structures of biological and abiological origins. Recent advances in computational tools, particularly the ones involving modern deep learning algorithms, have been shown to complement experimental approaches for describing and rationalizing PPIs. However, most of the existing works on PPI predictions use protein-sequence information, and thus have difficulties in accounting for the three-dimensional organization of the protein chains.
Results
In this study, we address this problem and describe a PPI analysis based on a graph attention network, named
Struct2Graph
, for identifying PPIs directly from the structural data of folded protein globules. Our method is capable of predicting the PPI with an accuracy of 98.89% on the balanced set consisting of an equal number of positive and negative pairs. On the unbalanced set with the ratio of 1:10 between positive and negative pairs, Struct2Graph achieves a fivefold cross validation average accuracy of 99.42%. Moreover, Struct2Graph can potentially identify residues that likely contribute to the formation of the protein–protein complex. The identification of important residues is tested for two different interaction types: (a) Proteins with multiple ligands competing for the same binding area, (b) Dynamic protein–protein adhesion interaction. Struct2Graph identifies interacting residues with 30% sensitivity, 89% specificity, and 87% accuracy.
Conclusions
In this manuscript, we address the problem of prediction of PPIs using a first of its kind, 3D-structure-based graph attention network (code available at
https://github.com/baranwa2/Struct2Graph
). Furthermore, the novel mutual attention mechanism provides insights into likely interaction sites through its unsupervised knowledge selection process. This study demonstrates that a relatively low-dimensional feature embedding learned from graph structures of individual proteins outperforms other modern machine learning classifiers based on global protein features. In addition, through the analysis of single amino acid variations, the attention mechanism shows preference for disease-causing residue variations over benign polymorphisms, demonstrating that it is not limited to interface residues.
Journal Article