Suchergebnisse

AbstractMany compelling examples have recently been provided in which people can achieve impressive epistemic success, e.g. draw highly accurate inferences, by using simple heuristics and very little information. This is possible by taking advantage of the features of the environment. The examples suggest an easy and appealing naturalization of rationality: on the one hand, people clearlycanapply simple heuristics, and on the other hand, they intuitivelyoughtdo so when this brings them high accuracy at little cost.. The 'ought-can' principle is satisfied, and rationality is meaningfully normative. We show, however, that this naturalization program is endangered by a computational wrinkle in the adaptation process taken to be responsible for this heuristics-based ('ecological') rationality: for the adaptation process to guarantee even minimal rationality, it requires astronomical computational resources, making the problem intractable. We consider various plausible auxiliary assumptions in attempt to remove this obstacle, and show that they do not succeed; intractability is a robust property of adaptation. We discuss the implications of our findings for the project of naturalizing rationality.

"Fast and Frugal Heuristics" published on by Oxford University Press.

In this work, we address the problem of airborne contaminant sensor placement in high-risk buildings where critical infrastructures are managed and operated, making them possible locations for terrorist attacks (such as governmental buildings and ministries, utilities, airports and hospitals). A new software is presented based on the "Matlab-CONTAM Toolbox" and the CONTAM multi-zone simulation software, to construct multiple scenarios of contamination events and to solve the multi-objective sensor placement problem for minimizing the average and maximum impact risk with respect to the contaminant mass inhaled impact metric. The use of the software is demonstrated in a case-study using the Holmes's House benchmark. The Toolbox is released under an open-source license at https://github.com/KIOS-Research/ matlab-contam-toolbox . This research work has been partially funded by the European Research Council under the ERC Advanced Grant ERC-2011-ADG-291508 "Fault-Adaptive Monitoring and Control of Complex Distributed Dynamical Systems" (FAULT-ADAPTIVE) and by the EC project HOME/2011/CIPS/AG "online identification of Failure and Attack on interdependent Critical InfrastructurES" (FACIES).

Cover -- Half Title -- Title Page -- Copyright Page -- Table of Contents -- Preface -- 1 How We Define Machine Learning -- 2 Preliminaries -- 2.1 Data and Dataset Types -- 2.1.1 Cross-Sectional -- 2.1.2 Time-Series -- 2.1.3 Panel -- 2.2 Plots -- 2.3 Probability Distributions with R -- 2.4 Regressions -- 2.4.1 Ordinary Least Squares (OLS) -- 2.4.2 Maximum Likelihood Estimators -- 2.4.3 Estimating MLE with R -- 2.5 BLUE -- 2.6 Modeling the Data -- 2.7 Causal vs. Predictive Models -- 2.7.1 Causal Models -- 2.7.2 Predictive Models -- 2.8 Simulation -- Part 1 Formal Look at Prediction -- 3 Bias-Variance Tradeoff -- 3.1 Estimator and MSE -- 3.2 Prediction - MSPE -- 3.3 Biased Estimator as a Predictor -- 3.4 Dropping a Variable in a Regression -- 3.5 Uncertainty in Estimations and Predictions -- 3.6 Prediction Interval for Unbiased OLS Predictor -- 4 Overfitting -- Part 2 Nonparametric Estimations -- 5 Parametric Estimations -- 5.1 Linear Probability Models (LPM) -- 5.2 Logistic Regression -- 5.2.1 Estimating Logistic Regression -- 5.2.2 Cost Functions -- 5.2.3 Deviance -- 5.2.4 Predictive Accuracy -- 6 Nonparametric Estimations - Basics -- 6.1 Density Estimations -- 6.2 Kernel Regressions -- 6.3 Regression Splines -- 6.4 MARS - Multivariate Adaptive Regression Splines -- 6.5 GAM - Generalized Additive Model -- 7 Smoothing -- 7.1 Using Bins -- 7.2 Kernel Smoothing -- 7.3 Locally Weighted Regression loess() -- 7.4 Smooth Spline Regression -- 7.5 Multivariate Loess -- 8 Nonparametric Classifier - kNN -- 8.1 mnist Dataset -- 8.2 Linear Classifiers (again) -- 8.3 k-Nearest Neighbors -- 8.4 kNN with Caret -- 8.4.1 mnist_27 -- 8.4.2 Adult Dataset -- Part 3 Self-Learning -- 9 Hyperparameter Tuning -- 9.1 Training, Validation, and Test Datasets -- 9.2 Splitting the Data Randomly -- 9.3 k-Fold Cross-Validation -- 9.4 Cross-Validated Grid Search.

Selective pressures favoring rapid decisions would have led to the evolution of simple decision-making mechanisms that could take the form of heuristics and rules that use as little available information as possible. Such decision heuristics can only be ecologically rational—yielding accurate inferences in particular problem domains—if they exploit the way that information is structured in the environment. The author presents a variety of fast and frugal heuristics that are ecologically rational and shows how they can be organized in the mind's adaptive toolbox of decision-making strategies.

This article accounts for the EU's failure to implement a strategy of Conflict Transformation (CT) through Regional Integration (RI) in four selected regions: Latin America, Asia, Africa, and the Middle East and North Africa (MENA). As a normative power, the EU has traditionally acted as a promoter of regional integration, with an underlying view of reducing regional conflicts. The academic literature, however, has failed to assess the exact role of the 'RI-CT nexus'1 in the EU's foreign policy toolbox, as well as its outcomes. To fill this gap, our study investigates the degree to which the RI-CT strategy is perceived by EU practitioners as a key deliverable and what concrete measures or influence paths have been implemented by the EU. We show that the EU has made use of four main types of influence paths in order to transform regional conflict dynamics. Nevertheless, our findings also demonstrate that RI developments within the different regions and, in particular, negative regional responses, progressively led EU practitioners to adapt and develop a behaviour of self-restraint.

To understand mental self-government of the developing reading and writing brain, correlations of clustering coefficients on fMRI reading or writing tasks with BASC 2 Adaptivity ratings (time 1 only) or working memory components (time 1 before and time 2 after instruction previously shown to improve achievement and change magnitude of fMRI connectivity) were investigated in 39 students in grades 4 to 9 who varied along a continuum of reading and writing skills. A Philips 3T scanner measured connectivity during six leveled fMRI reading tasks (subword—letters and sounds, word—word-specific spellings or affixed words, syntax comprehension—with and without homonym foils or with and without affix foils, and text comprehension) and three fMRI writing tasks—writing next letter in alphabet, adding missing letter in word spelling, and planning for composing. The Brain Connectivity Toolbox generated clustering coefficients based on the cingulo-opercular (CO) network; after controlling for multiple comparisons and movement, significant fMRI connectivity clustering coefficients for CO were identified in 8 brain regions bilaterally (cingulate gyrus, superior frontal gyrus, middle frontal gyrus, inferior frontal gyrus, superior temporal gyrus, insula, cingulum—cingulate gyrus, and cingulum—hippocampus). BASC2 Parent Ratings for Adaptivity were correlated with CO clustering coefficients on three reading tasks (letter-sound, word affix judgments and sentence comprehension) and one writing task (writing next letter in alphabet). Before instruction, each behavioral working memory measure (phonology, orthography, morphology, and syntax coding, phonological and orthographic loops for integrating internal language and output codes, and supervisory focused and switching attention) correlated significantly with at least one CO clustering coefficient. After instruction, the patterning of correlations changed with new correlations emerging. Results show that the reading and writing brain's mental government, supported by both CO ...

Chapter 1 Introduction: The Goal and Contents of This Dissertation 1 ECOLOGICALLY RATIONAL HEURISTICS 2 ACT-R: A UNIFIED THEORY OF COGNITION 5 ORGANIZATION OF THE DISSERTATION 8 Chapter 2 Methodological Preliminaries 10 INTRODUCTION 10 OVERVIEW OF THE CHAPTER AND INTRODUCTORY DEFINITIONS 11 HOW TO SELECT BETWEEN FORMAL MODELS: A SHORT OVERVIEW 12 CONCLUSION 18 Chapter 3 How Memory Aids Strategy Selection 19 ABSTRACT 19 INTRODUCTION 19 HOW MEMORY AIDS STRATEGY SELECTION: THE HYPOTHESIS OF NON-OVERLAPPING COGNITIVE NICHES 21 JUDGMENTS FROM THE ACCESSIBILITY OF MEMORIES 22 THE ADAPTIVE TOOLBOX 23 OVERVIEW OF THE CHAPTER 25 AN ACT-R MODEL OF NON-OVERLAPPING COGNITIVE NICHES 26 OVERVIEW OF SIMULATION STUDIES 1 AND 2 33 HOW I CALIBRATED THE MODEL TO PREDICT MEMORY RETRIEVAL (SIMULATION STUDY 1) 33 TESTS OF THE TARTLE NICHE AND TARTLE–KNOWLEDGE NICHE HYPOTHESES (SIMULATION STUDY 2) 42 WHEN WOULD A PERSON USING THE FLUENCY HEURISTIC MAKE ACCURATE INFERENCES? (EXPERIMENTS 1–6; SIMULATION STUDIES 3, 4) 44 ACCURACY AS A DETERMINANT OF STRATEGY SELECTION (EXPERIMENT 7) 53 EFFORT AS A DETERMINANT OF STRATEGY SELECTION (EXPERIMENT 8) 58 TIME AS A DETERMINANT OF STRATEGY SELECTION (EXPERIMENT 9) 62 GENERAL DISCUSSION 67 CONCLUDING REMARKS: EMERGING COGNITIVE NICHES 72 Chapter 4 Models of Recognition-based Multi-alternative Inference 74 ABSTRACT 74 INTRODUCTION 74 GENERALIZING THE RECOGNITION HEURISTIC: ELIMINATION BY RECOGNITION 75 ALTERNATIVE MODELS OF INFERENCE: A COMPETITION 76 TOWARD A THEORY OF STRATEGY SELECTION BY DEFAULT 79 OVERVIEW OF THE EXPERIMENTS: RECOGNITION IN POLITICAL ELECTIONS 82 DO CONFLICTING CUES OVERRULE THE RELIANCE ON RECOGNITION? (EXPERIMENT 11; REANALYSIS OF EARLIER ELECTION STUDY) 83 GENERALIZING THE RECOGNITION HEURISTIC: HOW EPISODIC KNOWLEDGE AIDS USING RECOGNITION (EXPERIMENT 12) 89 IS A LACK OF KNOWLEDGE INFORMATIVE ABOUT THE PREDICTIVE ACCURACY OF RECOGNITION? (EXPERIMENT 13) 98 HOW WELL DO COMPENSATORY MODELS PREDICT INDIVIDUAL INFERENCES? (EXPERIMENT 14; REANALYSIS OF EXPERIMENT 7) 102 DO ...

This thesis explores the role of adaptive learning in explaining empirical phenomena such as the Great Inflation. We show that the historic monetary policy deviated from the optimally recommended policy under learning. These deviations are caused by Brainard-type uncertainty which induces a conservative policy stance. Accounting for both imperfect knowledge and estimation uncertainty, the optimal policy under adaptive learning is consistent with historic policy behavior. However, we show that an optimizing but informationally constrained policy maker would most likely have experienced a Great Inflation like episode. Furthermore, we develop an estimation method to implement the adaptive learning hypothesis into DSGE modelling, thus extending the standard estimation toolbox.

Wettbewerbsfähig bleiben und die Erfolgsfaktoren Agilität und Selbstorganisation implementieren! Anpassungsfähigkeit, Schnelligkeit, volatile Märkte – nahezu alle Unternehmen müssen sich mit diesen Anforderungen auseinandersetzen – auch größere Unternehmen und Konzerne. Agile Teams alleine sind jedoch keine Antwort, sondern es muss über alle Abteilungen hinweg eine Veränderung in Richtung Agilität und Selbstorganisation stattfinden. Eine kaum lösbare Aufgabe für Unternehmen mit traditionellen Strukturen und Mitarbeitenden, die jahrelang, oftmals jahrzehntelang anders sozialisiert wurden und an ihrer bisherigen Unternehmenskultur festhalten wollen. Dieses Buch zeigt, wie diese Transformation auch in größeren Unternehmen gelingen kann. Wo und wie anfangen? Wie können Hierarchien schrittweise aufgelöst werden? Wie übernehmen Mitarbeitende die Führungsaufgaben? Wie können alle Mitarbeitenden mitgenommen werden, auch die eher schwierigen oder die, deren Macht schmilzt? Welche Rolle spielen die Betriebsräte? Wie führt man so ein Unternehmen? Was sind adaptive Organisationsstrukturen? Das sind nur einige Fragen, die dieses konsequent auf die praktische Umsetzung ausgerichtete Werk beantwortet. - Schritt-für-Schritt-Umsetzungsanleitung und Erfahrungsberichte - Zeigt, wo sich Chancen, aber auch Herausforderungen verbergen - Viele reale Beispiele und praktische Toolbox - Mit Arbeitshilfen und weiteren Infos zum Download

The governance of major metropolitan areas is often associated with a "fragmented" and "uncoordinated" multi-government apparatus, frequently sculpted from years of particularistic ad hoc administrative reforms. This image of dysfunctional structure gains high salience when the metropolitan context is accentuated by complexity and fluidity, especially where intense paradoxical forces of economic development and ecological sustainability are present. The most visible "solutions" for such a state often come from bureaucrats seeking to "streamline" government according to norms of standardization and hierarchy. But, calls for reform may also come from scholars of polycentric government, who see the problem as a misalignment of administrative structure with the metropolitan context. This article adopts the latter, less-appreciated perspective that argues such dysfunctions in a metropolitan multi-government network are essentially problems of adaptive organizational design. Different than the bureaucratic model, treatises on new public management or group-behavior theory, it emphasizes the contextual nature of public administration by employing the holistic framework of "organizational systems." It illustrates the logic by introducing a toolbox for multi-government design that speaks to the adaptive qualities of government networks in whole metropolitan areas. Its purpose is to reinvigorate this holistic approach in thinking about the way we look at multi-government networks in major metropolitan areas.