Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

‘The Production of Space’, in: Frans Jacobi, Imagine, Space Poetry, Copenhagen, 1996, unpaginated.

The Production of Space

From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

Qualitative Interviewing: The Art of Hearing Data

A prevalent and pervasive view of designing is that it can be modeled using variables and decisions made about what values should be taken by these variables. The activity of designing is carried out with the expectation that the designed artifact will operate in the natural world and the social world. These worlds impose constraints on the variables and their values; so, design could be described as a goal-oriented, constrained, decision- making activity. However, design distinguish- es itself from other similarly described activities not only by its domain but also by additional necessary features. Designing involves exploration, exploring what variables might be appropriate. The process of explo- ration involves both goal variables and deci- sion variables. In addition, designing involves learning: Part of the exploration activity is learning about emerging features as a design proceeds. Finally, design activity occurs within two contexts: the context within which the designer operates and the context produced by the developing design itself. The designer’s perception of what the context is affects the implication of the context on the design. The context shifts as the designer’s perceptions change. Design activity can be now characterized as a goal-oriented, con- strained, decision-making, exploration, and learning activity that operates within a con- text that depends on the designer’s percep- tion of the context.

Design Prototypes: A Knowledge Representation Schema for Design

The justified plan graph (JPG) was the first practical analytical method developed as part of the theory of Space Syntax, which purported to provide a graphical, mathematical and associated theoretical model for analysing the spatial configuration of buildings. In spite of early interest, in recent years relatively little research using this method has been published, perhaps because the JPG method is rarely explained in its totality and when it is, the descriptions are often inconsistent or unclear. Although it is now embedded in several software programs and its use may be more widespread, it is no better understood and after processing there is a marked lack of consistency in how the results are interpreted. This paper provides a historical background for the development of the JPG and a discussion of its conceptual or theoretical origins, followed by a “worked example” of the mathematics of the JPG. In combination with the results for two further cases, the paper identifies some important interpretative limits in the method and uses the examples to explain its potential use in design analysis. Finally, the paper discusses how the consistent application of this method to sets of related buildings is likely to produce a more valuable, and statistically viable, basis for future work.

/pdf/the-mathematics-of-spatial-configuration-revisiting-revising-s992bbm1xs.pdf

The Mathematics of Spatial Configuration: Revisiting, Revising and Critiquing Justified Plan Graph Theory

Michael Ostwald examines the intricate, constantly shifting relationship between architecture and fractal geometry. At times this dependence is diffuse, and modes of theoretical transference are subtle, symbolic or semiological. At other times wholesale appropriations of geometry takes place and large fragments of theory are pirated away from their originating discipline and used opportunistically.

/pdf/fractal-architecture-late-twentieth-century-connections-4ss70b9c14.pdf

“Fractal Architecture”: Late Twentieth Century Connections Between Architecture and Fractal Geometry

Researchers in the field of spatial psychology and environmental preference theory have tested a range of claims about the capacity of certain spatial configurations to evoke a positive sense of wellbeing in observers. In parallel, across the landscape, urban, architectural and interior design disciplines, there has been a growing acceptance that a balance of spatial characteristics—including prospect, refuge, mystery and complexity—is desirable in a natural, urban or interior environment. Yet, the evidence that the design disciplines cite for the desirability of these characteristics is often entirely qualitative and only rarely acknowledges the results from the fields of spatial psychology and environmental preference theory. The purpose of this paper is to provide a critical overview of the results of quantitative research which has been undertaken into the veracity of prospect-refuge theory and closely associated aspects of environmental preference theory. This meta-analysis not only involves a review of the results, but also their broad classification to develop a more holistic picture of the field, its findings and any gaps. The purpose of this process is not, explicitly at least, to assess the believability or rigour of this past research, but rather to examine and classify the findings, both for and against prospect-refuge theory, in a way that is useful for the design disciplines. Urban and interior studies supported the significance of prospect, and were more neutral about refuge. Studies related to natural environments provided evidence for the significance of both prospect and refuge, which has been linked to comfort, but also included evidence against and a neutral finding. More specifically for designers, the results for complexity seem to confirm that a degree of complexity in interior space is preferred, but they are unclear about how much or where it should be. The results for mystery are less emphatic with the majority being neutral or contrary. The quantitative evidence for prospect-refuge theory remains inconsistent. It is especially problematic that the results which are most commonly cited in architecture relate to studies of natural environments, not interiors or urban environments. As this paper demonstrates, the results are most valid in specific venues.

/pdf/evidence-for-prospect-refuge-theory-a-meta-analysis-of-the-28e7wbbi57.pdf

Evidence for prospect-refuge theory: a meta-analysis of the findings of environmental preference research

Understanding architectural education in Australasia

The Justified Plan Graph (JPG) technique was developed in the late 1970s and refined in the following two decades as a means of undertaking qualitative and quantitative research into the spatial structure or permeability of buildings. Famously used by Space Syntax researchers to uncover the social logic of architectural types, the technique remains an important, if not widely understood, approach to the analysis of the built environment. This paper uses the JPG method to undertake a three-stage analysis of the early houses of Pritzker Prize winning architect Glenn Murcutt; the stages are visual analysis, mathematical analysis and theoretical analysis. Through this process the paper offers a rare application of the JPG method to multiple works by the same architect and demonstrates the construction of a series of “inequality genotypes”, a partial “statistical genotype” and a partial “statistical archetype” for these houses. Instead of seeking to uncover the social structure of Murcutt’s housing, the paper analyses the architect’s distinctive approach to ordering space within otherwise simple volumes or forms. The ultimate purpose of this analysis is to offer an alternative space-based, rather than form-based, insight into this architect’s work.

/pdf/a-justified-plan-graph-analysis-of-the-early-houses-1975-4xsz0l9rcl.pdf

Michael J. Ostwald

Papers

The Mathematics of Spatial Configuration: Revisiting, Revising and Critiquing Justified Plan Graph Theory

“Fractal Architecture”: Late Twentieth Century Connections Between Architecture and Fractal Geometry

Evidence for prospect-refuge theory: a meta-analysis of the findings of environmental preference research

Understanding architectural education in Australasia

A Justified Plan Graph Analysis of the Early Houses (1975-1982) of Glenn Murcutt