Machine learning and the city : applications in architecture and urban design / edited by Silvio Carta.

Includes bibliographical references and index.

Contents

Cover
Title page
Copyright
Preface
Acknowledgements
Introduction
Section I Urban Complexity
1 Urban Complexity
2 Emergence and Universal Computation
3 Fractals and Geography
Project 1 Emergence and Urban Analysis
Project 2 The Evolution and Complexity of Urban Street Networks
Section II Machines that Think
4 Artificial Intelligence, Logic, and Formalising Common Sense
5 Defining Artificial Intelligence
6 AI: From Copy of Human Brain to Independent Learner
7 The History of Machine Learning and Its Convergent Trajectory Towards AI
8 Machine Behaviour
Project 3 Plan Generation from Program Graph
Project 4 Self-organising Floor Plans in Care Homes
Project 5 N2P2 – Neural Networks and Public Places
Project 6 Urban Fictions
Project 7 Latent Typologies: Architecture in Latent Space
Project 8 Enabling Alternative Architectures
Project 9 Distant Readings of Architecture: A Machine View of the City
Section III How Machines Learn
9 What Is Machine Learning?
10 Machine Learning: An Applied Mathematics Introduction
11 Machine Learning for Urban Computing
12 Autonomous Artificial Intelligent Agents
Project 10 Machine Learning for Spatial and Visual Connectivity
Project 11 Navigating Indoor Spaces Using Machine Learning: Train Stations in Paris
Project 12 Evolutionary Design Optimisation of Traffic Signals Applied to Quito City
Project 13 Constructing Agency: Self-directed Robotic Environments
Section IV Application to the City
13 Code and the Transduction of Space
14 Augmented Reality in Urban Places: Contested Content and the Duplicity of Code
15 Spatial Data in Urban Informatics: Contentions of the Software-sorted City
16 Urban Morphology Meets Deep Learning: Exploring Urban Forms in One Million Cities, Towns, and Villages Across the Planet
17 Computational Urban Design: Methods and Case Studies
18 Indexical Cities: Personal City Models with Data as Infrastructure
19 Machine Learning, Artificial Intelligence, and Urban Assemblages
20 Making a Smart City Legible
Project 14 A Tale of Many Cities: Universal Patterns in Human Urban Mobility
Project 15 Using Cellular Automata for Parking Recommendations in Smart Environments
Project 16 Gan Hadid
Project 17 Collective Design for Collective Living
Project 18 Architectural Machine Translation
Project 19 Large-scale Evaluation of the Urban Street View with Deep Learning Method
Project 20 Urban Portraits
Project 21 ML-City
Project 22 Imaging Place Using Generative Adversarial Networks (GAN Loci)
Project 23 Urban Forestry Science
Section V Machine Learning and Humans
21 Ten Simple Rules for Responsible Big Data Research
22 A Unified Framework of Five Principles for AI in Society
23 The Big Data Divide and Its Consequences
24 Design Fiction: A Short Essay on Design, Science, Fact, and Fiction
25 Superintelligence and Singularity
26 The Social Life of Robots: The Politics of Algorithms, Governance, and Sovereignty
Project 24 Experiments in Synthetic Data
Project 25 Emotional AI in Cities: Cross-cultural Lessons from the UK and Japan on Designing for an Ethical Life
Project 26 Decoding Urban Inequality: The Applications of Machine Learning for Mapping Inequality in Cities of the Global South
Project 27 Amsterdam 2040
Project 28 Committee of Infrastructure
Index
End User License Agreement

List of Figures

Chapter 1
Figure 1.1 A simulated random walk...
Figure 1.2 Building footprints...
Figure 1.3 Differences in road morphology...
Chapter 3
Figure 3.1 FrAU1: The reference...
Figure 3.2 Fractal models...
Figure 3.7 Simulation of urban...
Figure 3.3 Fractal analysis...
Figure 3.4 Radial analysis...
Figure 3.5 Evolution of the correlation...
Figure 3.6 Multifractal analysis...
Chapter 3
Figure 3p1.1 City Analysis...
Figure 3p1.2 Analysis based...
Figure 3p1.3 Simulation development...
Figure 3p1.4 Tools and outputs...
Chapter 3
Figure 3p2.1 The location of the city...
Figure 3p2.2 Evolution of the street...
Figure 3p2.3 Evolution of the street...
Chapter 5
Figure 5.1 The Turing Test...
Figure 5.2 The basic structure...
Figure 5.3 Rosenblatt’s perceptron...
Figure 5.4 The scientific method.
Chapter 8
Figure 8.1 Questions of concern to machine...
Figure 8.2 Machine behaviour lies...
Figure 8.3 The four categories...
Figure 8.4 AI systems represent...
Chapter 8
Figure 8p3.1 Graph representation...
Figure 8p3.2 Message passing network.
Figure 8p3.3 Architecture of the GCN...
Figure 8p3.4 Testing on L1/L2 losses.
Chapter 8
Figure 8p4.1 Example of some...
Figure 8p4.2 The three steps...
Figure 8p4.3 An overview...
Figure 8p4.4 The final floor plan...
Figure 8p4.5 The final floor plan...
Chapter 8
Figure 8p5.1 Visualisation of some...
Figure 8p5.2 Visual representation...
Figure 8p5.3 Aerial view of Notre Dame...
Figure 8p5.4 Rendering of the cross-section...
Chapter 8
Figure 8p6.1 Urban Fictions: this city does not exist...
Figure 8p6.2 Urban Fictions: what happens when architects...
Figure 8p6.3 Urban Fictions: this city does...
Figure 8p6.4 Urban Fictions. This city does...
Chapter 8
Figure 8p7.1 Interpolations between two...
Figure 8p7.2 Interpolations between four...
Figure 8p7.3 Interpolations between four...
Chapter 8
Figure 8p8.1 Model output key.
Figure 8p8.2 Model training data layout.
Figure 8p8.3 Generated plans.
Figure 8p8.4 User interface.199
Chapter 8
Figure 8p9.1 A Machine View of Groningen...
Figure 8p9.2 Plot of 30 000 buildings...
Figure 8p9.3 A large-scale projected...
Figure 8p9.4 Plot of 30 000 buildings...
Chapter 10
Figure 10.1 Likelihood versus...
Figure 10.2 Log likelihood versus...
Figure 10.3 Log likelihood for...
Figure 10.4 (Mis)identification...
Figure 10.5 Receiver operating...
Figure 10.6 Fit a straight line through...
Figure 10.7 The logarithm of the...
Figure 10.8 The linear fit.
Figure 10.9 Illustrating gradient...
Figure 10.10 Error versus epochs...
Figure 10.11 Training and testing...
Figure 10.12 Looks like we...
Figure 10.13 Examples of low...
Figure 10.14 The true relationship...
Figure 10.15 Underfitted model with...
Figure 10.16 A complicated model...
Figure 10.17 The same model...
Figure 10.18 A good fit using...
Figure 10.19 Same model...
Figure 10.20 The various...
Figure 10.21 The contour...
Figure 10.22 The information...
Figure 10.23 Demonstration...
Chapter 12
Figure 12.1 The scheme of the evolutionary process.
Figure 12.2 Crossover operator examples.
Figure 12.3 The common mutation operators.
Figure 12.4 The NEAT genome mapping to the phenotype ANN.
Figure 12.5 The diagram of structural...
Figure 12.6 The recombination of parental...
Figure 12.7 The hard maze schema...
Figure 12.8 The scheme of sensors...
Figure 12.10 The simple maze...
Figure 12.9 The topology of the phenotype...
Figure 12.11 The topology of control ANN...
Figure 12.12 The route of the winner...
Figure 12.13 The final positions...
Figure 12.14 The topology of control...
Figure 12.15 The route of the winner...
Figure 12.16 The final positions...
Chapter 12
Figure 12p10.1 On the left, a black...
Figure 12p10.2 Data set used for training...
Figure 12p10.3 The first row shows...
Chapter 12
Figure 12p11.1 Diagram of methodology...
Figure 12p11.2 Profile view of the study...
Figure 12p11.3 Images of (a) similar and (b)...
Chapter 12
Figure 12p12.1 Change in CO2 emissions.
Figure 12p12.2 Evolutionary algorithm flow chart.
Figure 12p12.3 Simulation and EA integration.
Figure 12p12.4 Geolocation of signal...
Chapter 12
Figure 12p13.1 Hexifinity: kinetic transformations...
Figure 12p13.2 Interlace: furniture groups...
Figure 12p13.3 Coralations: the design uses...
Figure 12p13.4 Vortexture: the project incorporates...
Chapter 14
Figure 14.1 A search for the Bronze...
Figure 14.2 Geotagged photographs...
Figure 14.3 Size of Wikipedia article...
Figure 14.4 Downtown Lexington viewed...
Figure 14.5 User‐generated information...
Chapter 16
Figure 16.1 Data from more than...
Figure 16.2 Images of urban forms...
Figure 16.3 Original input...
Figure 16.4 Finding the six...
Figure 16.5 An automatically generated...
Figure 16.6 Distribution of urban forms...
Figure 16.7 Global distribution...
Figure 16.8 The topology of...
Figure 16.9 Distribution of 33 875 urban...
Chapter 17
Figure 17.1 An example of our work at the urban...
Figure 17.2 The sequence in which a computational...
Figure 17.3 This example design space has a total...
Figure 17.4 A simple example of how different...
Figure 17.5 Analysing and interpreting the data...
Figure 17.6 Examples of the massing options that...
Figure 17.7 Example of a computational model...
Figure 17.8 The Hawaii design space in Scout...
Figure 17.9 Leeside was constructed by studying...
Figure 17.10 A series of questions guides...
Chapter 18
Figure 18.1 Data as a sea of entropy...
Figure 18.2 Conceptual diagrams...
Figure 18.3 Correspondence between...
Figure 18.4 Encoding of an image into...
Figure 18.5 Lexicon of placeness...
Figure 18.6 Unfolded lexemes...
Figure 18.7 Relationship between...
Figure 18.8 Concrete locality...
Figure 18.9 Encoding of localities...
Figure 18.10 Lexicon of eventness...
Figure 18.11 Projection table spaceness...
Figure 18.12 ‘Ways of talking.’...
Figure 18.13 Activated Lexemes of spaceness...
Figure 18.14 Models of northern...
Figure 18.15 Activated lexemes in spaceness...
Figure 18.16 Tokyoness in Seoul rendered...
Figure 18.17 Tokyoness in Tokyo rendered...
Figure 18.18 Tokyoness in Seoul rendered...
Figure 18.19 Tokyoness in Seoul...
Figure 18.20 Tokyoness in Bangkok rendered...
Figure 18.21 Tokyoness in Bangkok rendered...
Figure 18.22 Tokyoness in Vancouver rendered...
Figure 18.23 Tokyoness in Vancouver...
Figure 18.24 Bridgeness in Köln rendered...
Figure 18.25 Bridgeness in London rendered...
Figure 18.26 Bridgeness in London rendered...
Figure 18.27 Bridgeness in Auckland rendered...
Figure 18.28 Bridgeness in Auckland rendered...
Figure 18.29 Watersideness in Auckland rendered...
Figure 18.30 Watersideness in Auckland rendered...
Figure 18.31 Concertness in Auckland rendered...
Figure 18.32 Concertness in Auckland rendered...
Figure 18.33 Sunsetness in Auckland rendered...
Figure 18.34 Sunsetness in Auckland rendered...
Chapter 19
Figure 19.1 Performance indicators based...
Figure 19.2 Design and performance space...
Figure 19.3 Interface for a data- and...
Figure 19.4 (a) Comparative performance...
Chapter 20
Figure 20.1 AI CCTV and microphone network...
Figure 20.2 IOAIL Class Marks act as a traffic...
Figure 20.3 Roomba automatic vacuum cleaner...
Figure 20.4 IOAIL class markers on public...
Figure 20.5 Whilst public IOAIL signage...
Figure 20.6 Spectrum analysis and AI...
Figure 20.7 A cash machine near Market...
Figure 20.8 Photograph of live action...
Figure 20.9 Screen grabs from...
Chapter 20
Figure 20p14.1 Global movements...
Figure 20p14.2 Urban movements...
Figure 20p14.3 The probability is increasing...
Chapter 20
Figure 20p15.1 Rules for the ‘game...
Figure 20p15.2 Car society scenario.
Figure 20p15.3 The system structure diagram.
Chapter 20
Figure 20p16.1 Selection of DCGAN outputs...
Figure 20p16.2 Example of DCGAN outputs...
Figure 20p16.3 Example of DCGAN outputs...
Figure 20p16.4 Example of DCGAN outputs...
Chapter 20
Figure 20p17.1 The training set is built from...
Figure 20p17.2 Local stakeholders are invited...
Figure 20p17.3 Output of the triple-decker neural...
Figure 20p17.4 Facade elements are affiliated with...
Chapter 20
Figure 20p18.1 Camera input, translated to...
Figure 20p18.2 Geometric input, translated...
Figure 20p18.3 Geometric input, translated...
Figure 20p18.4 Camera input, translated...
Chapter 20
Figure 20p19.1 Rating on the construction...
Figure 20p19.2 Rating on the continuity...
Figure 20p19.3 The rating criteria.
Chapter 20
Figure 20p20.1 Boston skyline...
Figure 20p20.2 Boston skyline and...
Figure 20p20.3 Boston skyline and...
Figure 20p20.4 Boston skyline and...
Figure 20p20.5 Boston skyline and...
Figure 20p20.6 Boston skyline and...
Figure 20p20.7 Boston skyline and...
Chapter 20
Figure 20p21.1 Barcelona Los Angeles...
Figure 20p21.2 Barcelona Marrakesh. ...
Figure 20p21.3 Rural California Barc...
Figure 20p21.4 Los Angeles Marrakesh...
Chapter 20
Figure 20p22.1 Coordinate locations...
Figure 20p22.2 Results derived from...
Figure 20p22.3 Final synthetic images...
Figure 20p22.4 Seamless transitions...
Chapter 20
Figure 20p23.4 Municipality data set...
Figure 20p23.1 Environmental discomfort...
Figure 20p23.2 Trees and solar radiation...
Figure 20p23.3 Potential trees areas...
Chapter 22
Figure 22.1 AI Knowledge Map...
Figure 22.2 An ethical framework of the five...
Chapter 25
Figure 25.1 Linear versus exponential...
Figure 25.2 The six epochs of evolution...
Figure 25.3 Countdown to Singularity...
Figure 25.4 Linear view of evolution...
Figure 25.5 Fifteen views of evolution...
Figure 25.6 Canonical milestones based...
Figure 25.7 A mathematical singularity...
Chapter 26
Figure 2624.1 Variations of synthetic data sets...
Figure 2624.2 The set of combinations of the conditions...
Figure 2624.3 A real and a synthetic...
Figure 2624.4 A 2D representation of the high-dimensional...
Chapter 26
Figure 26p26.1 Model 1: Map of the living...
Figure 26p26.2 Model 2: Map of the residential...
Figure 26p26.3 Mapping the accuracy for Model...
Figure 26p26.4 Five examples of different...
Chapter 26
Figure 26p27.1 Amsterdam 2040: landscape view...
Figure 26p27.2 A girl picking up a pizza...
Figure 26p27.3 A boy fighting with his autonomous...
Chapter 26
Figure 26p28.1 A meeting stakeholder chart...
Figure 26p28.2 How the intersection will work...
Figure 26p28.3 Transcript excerpts created using...
Figure 26p28.4 PETA’s image-classification algorithm...

List of Tables

Chapter 12
Table 12.1 The direction of rangefinder...
Table 12.2 The FOV of the pie-slice...
Chapter 13
Table 13.1 Geodemographic details...
Chapter 20p14
Table_20p 14.1 A summary of statistics...
Chapter 22
Table 22.1 The five principles...

Machine Learning and the City: Applications in Architecture and Urban Design delivers a robust exploration of machine learning (ML) and artificial intelligence (AI) in the context of the built environment. Relevant contributions from leading scholars in their respective fields describe the ideas and techniques that underpin ML and AI, how to begin using ML and AI in urban design, and the likely impact of ML and AI on the future of city design and planning.

Each section couples theoretical and technical chapters, authoritative references, and concrete examples and projects that illustrate the efficacy and power of machine learning in urban design. The book also includes:

An introduction to the probabilistic logic that underpins machine learning
Comprehensive explorations of the applications of machine learning and artificial intelligence to urban environments
Practical discussions of the consequences of applied machine learning and the future of urban design

Perfect for designers approaching machine learning and AI for the first time, Machine Learning and the City: Applications in Architecture and Urban Design will also earn a place in the libraries of urban planners and engineers involved in urban design. Provided by publisher.