Content uploaded by Carsten Röcker
Author content
All content in this area was uploaded by Carsten Röcker on Sep 11, 2017
Content may be subject to copyright.
Reasoning Under Uncertainty: Towards
Collaborative Interactive Machine Learning
Sebastian Robert1(B
), Sebastian B¨uttner2, Carsten R¨ocker1,2,3,
and Andreas Holzinger3
1Fraunhofer-Institute of Optronics, System Technologies and Image Exploitation,
Application Center Industrial Automation (IOSB-INA), Lemgo, Germany
sebastian.robert@iosb-ina.fraunhofer.de
2Ostwestfalen-Lippe University of Applied Sciences, Lemgo, Germany
{sebastian.buettner,carsten.roecker}@hs-owl.de
3Holzinger Group, HCI-KDD, Institute for Medical Informatics,
Statistics and Documentation, Medical University Graz, Graz, Austria
a.holzinger@hci-kdd.org
Abstract. In this paper, we present the current state-of-the-art of deci-
sion making (DM) and machine learning (ML) and bridge the two research
domains to create an integrated approach of complex problem solving
based on human and computational agents. We present a novel classifica-
tion of ML, emphasizing the human-in-the-loop in interactive ML (iML)
and more specific on collaborative interactive ML (ciML), which we under-
stand as a deep integrated version of iML, where humans and algorithms
work hand in hand to solve complex problems. Both humans and com-
puters have specific strengths and weaknesses and integrating humans
into machine learning processes might be a very efficient way for tack-
ling problems. This approach bears immense research potential for various
domains, e.g., in health informatics or in industrial applications. We out-
line open questions and name future challenges that have to be addressed
by the research community to enable the use of collaborative interactive
machine learning for problem solving in a large scale.
Keywords: Decision making ·Reasoning ·Interactive machine learn-
ing ·Collaborative interactive machine learning
1 Introduction and Motivation
Disregarding the application domain, i.e. whether in the medical domain or in
the industrial context, current developments such as the rapidly growing com-
munication infrastructure, the internet of things and increasing processing power
with services and applications on top of those lead to massive amounts of data
and new possibilities. Traditional analytic tools are not well suited to capturing
the full value of “big data”. Instead ML is ideal for exploiting opportunities hid-
den in data. Highly complex small batch production and personalized medicine
(precision medicine [1]) are two of many possible target scenarios. Both depend
c
Springer International Publishing AG 2016
A. Holzinger (Ed.): ML for Health Informatics, LNAI 9605, pp. 357–376, 2016.
DOI: 10.1007/978-3-319-50478-0 18
358 S. Robert et al.
on computer-intensive data processing prior to its analysis and decision making
processes.
However, to handle and exploit the required data, besides computer algo-
rithms, human capabilities are strongly needed as well. For example, classical
logic in ML approaches permits only exact reasoning, i.e. if A is true then A
is non-false and if B is false then B is non-true. However, even though modern
sophisticated automatic ML approaches can hardly cope with such situations,
human agents can deal with such deficiencies.
Moreover, many ML approaches are based on normative models such as for-
mal probability theory and expected utility (EU) theory. EU theory accounts for
decision under uncertainty and is based on axioms of rational behavior described
by von Neumann and Morgenstern (1944) [2]. Based upon the fact that informa-
tion available in daily problem solving situations is most of the time imperfect,
imprecise and uncertain due to time pressure, disturbance of unknown factors
or randomness outcome of some attributes [3,4], the interaction between human
and computer has to be designed in an optimal way in order to realize the best
possible output. Given that, a combined approach of human and computer input
can be a sustainable approach for effectively revealing structural or temporal
patterns (“knowledge”) and make them accessible for decision making.
At this point, decision theory comes into play and helps us to deal with
bounded rationality and the problem of which questions to pose to human
experts and how to ask those questions. Therefore, new types of human-computer
interaction (HCI) will arise and shape the ecosystem of human, technology and
organization. In particular, adaptive decision support systems that help humans
to solve complex problems and make far-reaching decisions will play a central
role in future work places.
In this paper, we will focus on decision making under uncertainty and bridge
it to ML research and particularly to interactive ML. After discussing the state-
of-the-art in ML and decision making under uncertainty, we provide some prac-
tical aspects for the integration of both approaches. Finally, we discuss some
open questions and outline future research avenues.
2 Glossary and Key Terms
Bias refers to a systematic pattern of deviation from rationality in decision
making processes.
Bounded Rationality – introduced by Herbert A. Simon [5]–isusedtodenote
the type of rationality that people resort to when the environment in which they
operate is too complex relative to their limited mental abilities [6].
Decision Support Systems (DSS) are intended to assist decision makers in taking
full advantage of available information and are a central part of health informat-
ics [7] and industrial applications [8].
Decision Theory is concerned with goal-directed behaviour in the presence of
options [9]. While normative decision theory focuses on identifying the opti-
mal decision to make, assuming a fully-rational decision maker who is able to
Reasoning Under Uncertainty 359
compute with perfect accuracy, descriptive decision theory deals with questions
pertaining to how people actually behave in given choice situations. Prescriptive
decision theory is the logic consequence and tries to exploit some of the logical
consequences of normative theories and empirical findings of descriptive studies
to make better choices [10].
Expected Utility (EU) Theory consists of four axioms, that define a rational deci-
sion maker: completeness, transitivity, independence, and continuity; if those are
satisfied, then the decision making is considered to be rational and the pref-
erences can be represented by a utility function, i.e. one can assign numbers
(utilities) [2].
Heuristics describe approaches to problem solving and decision making which
are not perfect, but sufficient for reaching immediate goals [11].
Human-Computer Interaction (HCI) is a multi-disciplinary research field that
deals with “the design, implementation and evaluation of interactive systems in
the context of the user’s task and work”[12, p. 4]. It can be located at the
intersection of psychology and cognitive science, ergonomics, computer science
and engineering, business, design, technical writing and other fields [12, p. 4].
Judgment and Decision Making (JDM) is a descriptive field of research which
focuses on understanding decision processes on an individual and group level.
Machine Learning (ML) is a research field grounding in computer science that
“concentrates on induction algorithms and on other algorithms that can be said
to ‘learn’”[13]. While in automatic Machine Learning (aML) representations of
real-world objects and knowledge are automatically generated from data, inter-
active Machine Learning (iML) methods allow humans to interact with com-
puters in some way to generate knowledge and find an optimal solution for a
problem. More specifically, collaborative interactive Machine Learning (ciML)
is a form of iML, where at least one human is integrated into the algorithm
using a specific user interface that allows manipulating the algorithm and its
intermediate steps to find a good solution in a short time.
Perception-Based Classification (PBC) is a classification of data done by humans
based on their visual perception. In the context of ML, PBC has been introduced
byAnkerstetal.[14] who enabled users to interactively create decision trees.
PBC can be seen as one possible way of realizing iML.
Utility Theorem describes that a decision-maker faced with probabilistic (partic-
ularly when probabilities are distorted or unknown) outcomes of different choices
will behave as if she/he is maximizing the expected value [15]; this is the basis
for the expected utility theory.
3 State-of-the-Art
In this section, we will provide an overview of the current research regarding
two fields: First, we will investigate machine learning (ML) and focus especially
360 S. Robert et al.
on the advances in interactive machine learning (iML). Second, we will provide
an overview of the research on JDM under uncertainty. We will further focus on
bridging the research on human decision making and the research on iML. We
will motivate, why the knowledge of and research on human decision-making is
key for the development of future human-oriented ciML systems.
3.1 Machine Learning (ML)
ML is a very practical field with many application areas, though at the same
time well grounded theories with many open research challenges exist. There
are many various definitions, depending on whom to ask; a Bayesian will give
a different answer than a Symbolist [16]; a classical definition is close to and
grounding in computer science that “concentrates on induction algorithms and
on other algorithms that can be said to ‘learn’”[13]. This definition is at the same
time the goal of ML which concentrates on the development of “programs that
automatically improve with experience”[17]. Advances in ML have solved many
practical problems, e.g., recognizing speech [18], giving movie recommendations
based on personal references [19] or driving a vehicle autonomously [20].
In the following, we will differentiate between classical ML approaches, that
we will call aML and the newer concepts of iML.
Automatic Machine Learning (aML): Methods and algorithms of machine
learning are often categorized as follows (here the classification of Marsland
[21]):
– With supervised learning methods, an algorithm creates a general model from
a training set of examples containing input and output data (targets). With
this model, the output of new unknown input can be predicted.
– Contrary, when using unsupervised learning methods, the output data are
not provided to the algorithm. The algorithm focuses on finding similarities
between a set of input data and classifies the data into categories.
–Reinforcement learning is somehow between supervised and unsupervised
learning. It characterizes algorithms that receive feedback, in the case that
their created output data are wrong. By this feedback the algorithm can
explore possibilities and iteratively find better models, respectively outputs.
– Finally evolutionary learning methods develop models iteratively by receiving
an assessment of the quality (fitness) of the current model. As the term depicts,
this learning method is inspired by the biological evolution.
The mentioned methods and algorithms all have in common, that they – once
started – run automatically. We therefore call those classical machine learning
methods automatic machine learning. When using aML methods, human involve-
ment is in general very limited and restricted to the following three aspects:
– Humans have to prepare the data and remove corrupt or wrong data sets from
the input data (data cleansing).
Reasoning Under Uncertainty 361
– When using supervised learning methods, humans are responsible for provid-
ing the output data, e.g., for labeling data in classification tasks.
– Another user involvement is the assessment of a certain model and the eval-
uation. Humans can assess the generated model and its results, and decide,
whether a certain model is able to produce good predictions or not.
The traditional approach does not put much emphasize on the human inter-
actions with the ML system. Humans are somehow involved in providing the data
as described above, but the early ML research mostly neglects the question, how
humans can provide data and how they deal with an inaccurate model. From
a practical perspective, this is a huge restriction in automatic machine learning
(aML) systems. The main problems of practical ML applications are often not
the implementation of the algorithm itself, but rather the data acquisition and
cleansing. Often data are corrupt or of bad quality and in most cases data do
not cover all required context information to solve a specific problem [3,4].
Interactive Machine Learning (iML): Compared to aML, iML is a relative
new approach that also considers the human involvement and interactions in
ML and aims at putting the human into the loop of machine learning. In this
section, we will discuss the approaches and concepts that previously have been
described under the term iML. We will distinguish in this section between three
types of iML methods: First, early works in the iML research considered iML
as an alternative way of ML where humans accomplish the model generation,
which basically means that humans replace algorithms. Second, concepts have
been proposed under the term iML that put a human into the training-evaluation
loop, but still execute algorithms automatically. Contrary to aML in this type of
iML algorithms have to be much faster to give rapid feedback to a user. Third,
humans can work hand in hand with algorithms to create a certain model, which
we consider as the most promising concept of iML with the best integration of
users and algorithms.
Humans replacing algorithms: Early work in iML has been done by Ankerst et al.
[14]. They implemented a system called perception-based classification (PBC)
that provides users the means to interactively create decision trees by visual-
izing the training data in a suitable way. By interacting with the visualized
training data, users select attributes and split points to construct the decision
trees. The system cannot automatically generate the tree. Instead, the user of
the system replaces the algorithm and creates the tree manually with the inter-
active application provided. According to their evaluation, the system reaches
the same accuracy as algorithmic classifiers but the human-generated decision
trees have a smaller tree size, which is beneficial in terms of understandability.
Another advantage of the interactive and manual approach is the possibility
of backtracking in case of a suboptimal subtree – a situation that humans can
easily recognize [14]. A huge benefit of this human-centered approach is the inte-
gration of the users’ domain knowledge into the decision tree construction [22].
362 S. Robert et al.
Building on the work of Ankerst et al., Ware et al. [23] developed a similar sys-
tem that replaces the algorithm with users. Their work focuses mainly on an
empirical evaluation of the performance of humans compared to state-of-the-art
algorithms. According to their study, novice users can build trees that are as
accurate as the ones provided from algorithms, but similar to Ankerst et al.
they found, that the tree size is decreased, when humans generate the decision
trees. On the other hand, Ware et al. point out that this manual iML approach
might not be suitable for large data sets and high-dimensional data. This early
variant of interactive machine learning is shown in Fig. 1A.
Humans in the training-evaluation loop: Another variety of iML is the integra-
tion of humans into the training-evaluation loop, when using supervised learning
methods. Fails and Olsen [24] were one of the first, who used the term iML and
proposed this integration for the rapid development of models, if the feature
selection cannot be done by domain-experts due to missing knowledge. They
give an example of the use of iML for the rapid development of perceptual user
interfaces (PUIs), that are developed by interaction designers who are usually
not familiar with computer vision (CV) algorithms. For this purpose, they pro-
vide a tool that gives designers rapid visual feedback of the produced classifiers
and the iterative changes of the selected features for the model generation. The
tool masks the complexity of the feature selection and rather allows users to
assess the output of the model generation and to drive the feature selection
into the right direction. A similar concept has been described by Fiebrink et al.
[25]. They developed Wekinator1, a system that analyses human gestures in the
context of music making. A graphical user interface supports users with the cre-
ation of appropriate training data, the configuration of various ML algorithms
and parameters and allows a real-time evaluation of the trained model by giving
visual or auditory feedback. This real-time evaluation allows a domain user to
rapidly adapt the input data to improve the model. Fogarty et al. [26]presented
CueFlik, a similar iML tool for generating models for image classification tasks.
For the mentioned type of iML, it is essential to have algorithms that have a
very short learning time to be able to give rapid feedback on the results [24].
Addressing this particular aspect in connection with big data, Simard et al. [27]
described a system that is very generic in terms of the data types and tasks and
interactive even when using big data. Their system called ICE (interactive clas-
sification and extraction) allows users to interactively build models consisting
of several millions of items. In [28] they extend their approach and addition-
ally deliver feedback about the performance of the generated model to the user.
With this system they empower users to not only optimize the model in terms
of accuracy, but to optimize in terms of performance as well. While the men-
tioned systems use only one model, in recent years model ensembles became the
standard of ML [16]. Talbort et al. therefore provide a tool that deals with mul-
tiple models and allows users to interactively build combination models [29]. All
mentioned publications in this section use the term iML to describe a concept,
1http://www.wekinator.org/.
Reasoning Under Uncertainty 363
Fig. 1. Classification of interactive machine learning (iML). A: Early iML research
aimed at replacing algorithms and using human pattern recognition capabilities instead.
B: Later iML methods have been proposed that provide a rapid feedback cycle to users.
Models are generated in a very short time and presented to users. Based on the pre-
sented model, users can adapt the input data and rerun the machine learning algorithm.
With this approach the model is iteratively improved. C: Using collaborative interac-
tive machine learning (ciML) humans can manipulate an algorithm during runtime
and improve the model while it is generated. Human and computational agents work
collaboratively on a specific problem.
where humans are in the training-evaluation loop, but cannot interfere with the
algorithm itself – from a human perspective the algorithm is a black-box. The
method of putting humans into the training-evaluation loop is shown in Fig. 1B.
Humans collaborating with algorithms: Sinard et al. define iML as a ML scenario,
where “the teacher can provide [...] information to the machine as the learning
ta sk p rogre sse s ”[27]. De facto, most systems presented in the past realized this
iML by providing means to users to evaluate a certain model and by changing
the training data to optimize the previously generated model. In this section,
364 S. Robert et al.
we present work that goes even one step further and integrates humans into
the process by providing a user interface that allows humans to manipulate the
parameters of the algorithm during its execution. We will call this approach col-
laborative interactive machine learning (ciML). Im this approach, humans can
directly collaborate with an algorithm. With this deep integration, new possi-
bilities of human-computer collaboration in ML might rise. One of the earliest
works, that aimed at the collaboration between human and algorithm in a ML
scenario has been presented by Ankerst et al. [30]. They built up on their earlier
PBCsystem[14] and provide an iML system for building decision trees for a
classification task. While their earlier PBC system only visualized data and left
the decision tree building to the users, algorithms are now integrated into the
system that might (but does not have to) be used. With the options provided,
different types of cooperation can be realized: manual (equivalent to the earlier
PBC), combined or completely automatic model generation. For the decision
tree construction, the system supports with proposing splits, with visualizing
hypothetical splits – up to a defined number of levels (“look-ahead function”),
and with the feature of automatically expanding subtrees. One mentioned goal of
their work is the use of human pattern recognition capabilities in the interactive
decision tree construction by still using algorithmic operations to allow dealing
with huge data sets [30]. Along these lines, Holzinger defines iML as “algorithms
that can interact with agents and can optimize their learning behavior through
these interactions, where the agents can also be human”[31], consequently, he
considers iML as this deeply integrated type of a collaboration between algo-
rithm and human. He discusses another issue that can be addressed with this
deeply integrated form of iML: Sometimes ML needs to deal with rare events,
like occurrences of rare diseases in health informatics, and consequently adequate
training data are missing. He identifies new application areas for ciML within the
health domain, e.g. for subspace clustering, protein folding, or k-anonymization
of patient data and names challenges for the future ciML research. Holzinger
also shows that the solution of complex problems is possible by using ciML. He
presents the integration of users into an ant colony algorithm to solve a traveling
salesman problem (TSP) [32]. A visualization shows the pheromone tracks of the
ants in the TSP and the optimal round-trip found by the algorithm so far. Users
can select edges and add or remove the current amount of pheromones on the
edge between each of the iterations. First experiments show that the process is
sped up in terms of required iterations to find the optimal solution [32]. The
collaborative variant of interactive machine learning is shown in Fig.1C. As the
related work regarding the collaboration between humans and algorithms in iML
shows, there has not been done a lot of research investigating the challenges and
opportunities of a human-algorithm interaction. Application areas of this new
iML approach need to be further identified and the implications of a human
agent in the iML system need to be explored. While humans can bring tacit
knowledge and context information into the process of building models, the
question remains unclear how human decisions effect the output of the iML sys-
tem. However, there has been a lot of research regarding human-decision making
that we will introduce in the next section.
Reasoning Under Uncertainty 365
3.2 Judgement and Decision Research
Generally, the main focus of ML is on dealing with uncertainty and making pre-
dictions. In order to infer unknowns, data sets have to be learned and analysed.
Therefore, most ML approaches are based on normative models such as formal
probability theory and EU theory. EU theory accounts for decisions under uncer-
tainty and is based on axioms of rational behavior, codified by von Neumann and
Morgenstern [2]. It states that the overall utility of an option equals the expected
utility, calculated by multiplying the utility and probability of each outcome [33,
p. 24]. Probability theory in ML is most often used in terms of Bayesian decision
theory [34–37], which is build on EU theory as a framework for solving problems
under uncertainty [38, p. 140]. “Individuals who follow these theories are said to
be rational” [39, p. 724].
The successful integration of knowledge of a domain expert in the black-box
as discussed in the iML approach stands or falls with the careful consideration
of people’s actual decision making abilities. It is generally accepted that human
reasoning and decision making abilities can exhibit various shortcomings when
compared with mathematical logic [3]. Hence, the question that arises is, how to
integrate human and computer input, accounting for the imperfections of both
[40, p. 2122]. At this point descriptive decision theory can offer useful insights
for the optimal integration of human judgement in iML approaches.
Descriptive decision theory deals with questions pertaining to how people
behave in given choice situations and what we need to fully predict their behav-
iour in such situations [41, p. 2]. In many cases, this is a difficult task due
to given inconsistencies in people’s choices. These inconsistencies can often be
attributed to irrational behaviour or accidental errors, which can also lead to
deficient decisions [41, p. 6].
Within the last decades, a growing research community within the area of
descriptive decision making is focusing on understanding individual and group
judgement and decision making (JDM) [42,43].2Researchers from various fields
are actively contributing to JDM, e.g. cognitive psychologists, social psycholo-
gists, statisticians and economists [42,45]. They have developed a detailed pic-
ture of the ways in which individuals judgement is bounded [46], e.g., people
violate the axions of EU theory and do not always follow basic principles of
calculus [47,48]. JDM tasks are characterized by uncertainty and/or by a con-
cern for individual’s preferences and will therefore apply to central aspects of
human activities in iML [38, p. 140]. In detail, JDM research focuses on how
different factors (e.g., information visualization) affect decision quality and how
it can be improved [49,50]. In order to give any predictions about human judge-
ment, JDM usually presupposes a definition of rationality that makes certain
actions measurable. This instrumental view of rationality only accords with nor-
mative theory if keeping in line with it helps to attain satisfaction – measured
in subjective utility [51]. A basic approach of JDM is to compare actual judge-
ments to normative models and look for deviations. These so called biases are
2See also [44] for the chapter.
366 S. Robert et al.
the starting point for building models that explain and predict human decision
making behaviour. A fundamental outcome of early JDM research reveals that
the typical model of a “rational man” as presumed by most normative theories –
considering every possible action, every outcome in every possible state and cal-
culating the choice that would lead to the best outcome – is unrealistic and does
not exist [5]. Instead innumerous studies revealed that people cannot carry out
the complex and time-consuming calculations necessary to determine the ideal
choice out of possible actions [52, p. 7]. Instead people act as “satisficers” and
make decisions on the basis of limited information, cognitive limitations and
the time available. Simon’s concept of bounded rationality describes how people
actually reach a judgement or a decision and has become a widely used model
for human decision behaviour [5].
Building on Simon’s model, Tversky and Kahneman developed their heuris-
tics and biases program that fundamentally shaped our understanding of judg-
ment as we know it today [48]. According to their argumentation, coming to a
decision requires a process of information search. Information can be retrieved
from memory or other external sources. In any case, information has to be pre-
processed for the particular problem and a final conclusion has so be drawn.
Therefore, information processing is key for decision making and limited cog-
nitive abilities, as stated in the model of bounded rationality, might essentially
impact decision quality. The major reason for the huge impact of the heuristics
and biases program in research is, that it is able to explain a wide variety of
different decision situations without restricting it due to motivated irrationality
[52, p. 1].
Tversky and Kahneman assume, that decisions under uncertainty are based
on heuristics rather than complex algorithms [48]. Heuristics are defined as men-
tal short-cuts or rules of thumb and require only limited amount of information
and cognitive abilities. Generally, heuristics achieve results fast and depend on
low effort. To do so, they neglect relevant information, which can lead to system-
atic predictable deviations from rationality. There is a huge amount of evidence
that biases can lead to poor outcomes in important and novel decisions [42,53].
This, together with the fact that biases are systematic, emphasises the impor-
tance of incorporating heuristics in modelling.
In their pioneering work, Tversky and Kahneman described three funda-
mental heuristics [48] which are relevant in countless practical situations. The
representativeness heuristic is applied when people make judgements about the
probability of an unknown event. To come up with a judgement, people tend to
judge the probability of the unknown event by finding a comparable known event
and assume that the probabilities will be similar. For illustration, Tversky and
Kahneman developed the “Linda problem”, where they describe the fictitious
person Linda as “31 years old, single, outspoken, and very bright. She majored
in philosophy. As a student, she was deeply concerned with issues of discrimi-
nation and social justice, and also participated in anti-nuclear demonstrations”
[54, p. 297]. Thereupon they asked subjects which is more probable, (a) Linda
being a bank teller or (b) Linda being a bank teller and actively involved in
Reasoning Under Uncertainty 367
feminist movement. Results reveal, that in accordance with their hypothesis, a
vast majority (80–90%) of subjects chose the conjunction (b) to be more likely
than the single event (a). From a logical perspective, a conjunction of events (b)
can never be more likely than any of its constituents (a) and therefore indicates
a violation of rationality. Within the last decades, many different biases have
been linked to the representativeness heuristic (e.g., conjunction fallacy, base
rate neglect, insensitivity to sample size) [42].
The availability heuristic is the second of Tversky and Kahnemans heuristics
and states, that people rely upon knowledge that is easily available and comes
to mind rather than complete data [55]. By relying on the availability of a given
event in someone’s memory, the actual probability of the event can often be
predicted quite good. Nevertheless, sometimes the availability of an event is
influenced by other factors besides the probability or frequency of the occurrence
and in this case the availability heuristic will lead to systematic deviations from
rationality [55]. For example the chronological distance or conciseness are factors
that can influence the availability of an event. The cause of death “firearm”
is estimated as much higher compared to “tobacco”, which can be attributed
to the media coverage of violence [42,56]. Similar to this, subjects who were
asked to estimate “If a random word is taken from an English text, is it more
likely that the word starts with a K, or that K is the third letter?” [55,p.
1125]. Following Tversky and Kahneman’s hypothesis, people easier recall words
beginning with an K and therefore overestimate the number of words that begin
with the letter K. Although experimental results support this hypothesis, a text
typically contains twice as many words which have the letter K at the third,
rather than first letter.
The so-called anchoring and adjustment heuristic describes a widely explored
and robust phenomenon in human decision making [48]. The heuristic can be
very useful when primary values of information do hint to a correct answer and
are relevant to the underlying decision-problem – a situation found in many daily
tasks. The anchor effect – as the central result of the anchoring and adjustment
heuristic – can be found in situations, where a numerical starting point (the
anchor) is processed to form a final estimation. In case the final estimation is
biased towards the initial starting point, one talks about an anchoring effect.
In a well-known demonstration, Tversky and Kahneman asked subjects to esti-
mate the percentage of African countries that are in the United Nations (UN)
[48, p. 1128]. Prior to this, for every subject of the experiment, a random num-
ber between one and one hundred after spinning a wheel of fortune was chosen.
Subjects had to state if the random number is higher or lower compared to
the true value. It was found, that people who had a lower number estimated
fewer countries in the UN than people who had a higher number. Thereupon
numerous experiments validated the robustness of the anchoring effect in vary-
ing fields of application, e.g. general knowledge [57], probability estimates [44,58]
and negotiations [59,60]. Neither financial incentives nor explicit advices could
effectivly mitigate the anchoring effect [61,62]. Moreover, the numerical start-
ing point does not have to be relevant to the underlying decision-problem, even
368 S. Robert et al.
unconsciously perceived or irrelevant values can distort the judgement [61,p.
123]. In general, there are two different approaches to explain the occurrence of
the anchor effect. The original approach of Tversky and Kahneman states that
individuals tend to anchor onto a numerical value and then gradually adjust
away from that value until they reach a decision that seems reasonable [48]. This
anchoring and adjustment process is usually insufficient and therefore biased. In
contrast, the selective accessibility approach argues, that biased estimations are
rooted in an early phase of information processing [57,63,64]. Following the
approach, individuals, when given an anchor, will evaluate the hypothesis that
the anchor is a suitable answer (confirmatory hypothesis testing) and therefore
access all the relevant attributes of the anchor value. Thereon, the approach
assumes that anchoring effects are mediated by the selectively increased acces-
sibility of anchor-consistent knowledge and the final estimate is therefore biased
towards the anchor. Overall, none of the mentioned approaches can fully explain
empirical evidence and the origin of the anchoring effect is still highly debated
within the research community [42,65].
In addition to the three fundamental heuristics and their resulting biases,
there are further heuristics which try to explain decision making under specific
situations. Despite the tremendous success of the heuristics and biases program,
there are alternative approaches to explain actual decision making behaviour. For
example the fast-and-frugal-approach – mostly based on Gigerenzers works – is
also based on several simple heuristics, but in contrast to the classical heuristics,
they are precisely defined and can be directly validated [66,67]. Moreover, the
probabilistic mental model [68] and prospect theory [69] also build on limited
cognitive abilities and are used in different areas to predict decision making
behaviour.
3.3 Practical Aspects for the Integration of Interactive ML
and Decision Theory
The importance of the integration of interactive ML and decision theory is evi-
dent. Given the massive consequences that can result from suboptimal decision
making, it is critical to improve our knowledge about ways to yield better deci-
sion outcomes [46, p. 379]. In our knowledge-based economy, each decision is
likely to have vast implications and will affect subsequent decisions on their
own. Decision problems have to be analysed for their potential receptiveness
to decision biases and in what ways they are likely to benefit from automatic
processing.
On the one side, current technological and methodical advances enable us to
cope with more complex decision tasks. But on the other side, in many practi-
cal situations decision making in terms of the interaction between human and
computer input is still limited and does not tap the full potential. Moreover,
new decision situations in many fields of application are characterized by the
same underlying process and therefore share the common need for new ways of
interaction.
Reasoning Under Uncertainty 369
For example, there are innumerable applications in the field of medical deci-
sion making and cyber-physical systems (e.g. “Industry 4.0”) such as assistance
or recommender systems that are based on the same abstract decision problem,
combine similar approaches of computer algorithms with human input and there-
fore face similar challenges. For instance, the analysis of sensor data is pretty
similar in many practical applications. On the one hand, data may describe
body parameters such as temperature, heartbeat or blood plasma concentration
in a medical context. On the other hand, data may provide information about
the energy consumption of a power unit, the temperature of an engine or the
status of a relay in an industrial context. Although there are many algorithms
that can analyse the captured data in a purely unsupervised fashion, in order
to achieve excellent and instant results, an interactive data analysis backed by
human decision making skills can offer new possibilities and bring context infor-
mation into the process. The same applies to the area of image exploitation. In
many cases, it is about finding structural anomalies in data and learning from
previous examples. With up-to-date methods of image exploitation, algorithms
can detect, count and cluster different types of objects. These algorithms are in
many cases only partially automatic and require human input. In medical image
exploitation, doctors can help to provide diagnostic findings in the segmentation
of skin cancer images [70]. In the industrial context, image exploitation is for
example used to detect tool wear [71]. In both situations, wrong diagnoses and
decisions potentially bear extensive risk and therefore the optimal integration
of human and computer input is of great importance. A big issue is accordingly
the integration process, because exactly here setting up a system between the
expert and the algorithm requires a common ground between them and is crucial
for total imaging. This common ground has to exploit computational power and
integrate human intelligence to realise the best possible output.
4 Open Problems
The study of ML is primarily based on normative models. Most of these models
are the result of centuries of refection and analysis and are widely accepted as
the basis of logical reasoning. For the fact that human decision making skills
are in certain settings superior to computer algorithms – e.g. many ML-methods
perform very badly on extrapolation problems which would be very easy for
humans [32, p. 4] – and major assumptions of normative models cannot be
applied in reality, a conjoint approach of human and machine input could be
key to enhanced decision quality. Therefore, the answer is to put humans in the
loop [40]. However, using normative models to integrate human decision making
in centrals parts of machine learning could lead to faulty predictions since the
nature of actual decision making is of bounded rationality [5].
Based on the described approaches, today we know the specific ways in which
decision makers are likely to be biased and we can describe how people make
decisions with astonishing detail and reliability. In addition, with regards to
normative models, we have a clear vision of how much better decision making
370 S. Robert et al.
could be [46]. The most important step now is to integrate those two different
approaches, correct biases and improve decision making. The prescriptions for
such corrections are called prescriptive models [33, p. 19] and will decide about
the success of human-in-the-loop approaches in ML. Altogether, not only do we
need to know the nature of the specific problem, “but normative models must
be understood in terms of their role in looking for biases, understanding these
biases in terms of descriptive models and developing prescriptive models” [72,
p. 20].
In consideration of this fact, interactive ML approaches are a promising can-
didate for further enhancing the knowledge discovery process. One important
problem which we have to face in future research is which questions to pose to
humans and how to ask those questions [40]. At this point, human machine-
interaction could provide useful insights and offer guidelines for the design of
interfaces and visualisations. Moreover, research in this area, i.e. at the intersec-
tion of cognitive science and computational science is fruitful for further improv-
ing ML thus improve performance on a wide range of tasks, including settings
which are difficult for humans to process (e.g., big data and high dimensional
problems) [32]. According to Lee and Holzinger [73], there is a very common
misconception about high dimensionality, i.e. that ML would produce better
outcomes with higher dimensional data. Increasing amounts of input features
can build more accurate predictors as features are key to learning and under-
standing. However, such attempts need high computational power, and due to
limitations in human perception, understanding structures in high dimensional
spaces is practically impossible. Hence, the outcome must be shaped in a form
perceivable for humans, which is a very difficult problem. Here graph-based rep-
resentations in R2are very helpful in that respect and open up a lot of future
possibilities [74,75].
5 Future Challenges
The important role of iML for dealing with complexity is evident. However,
future research has to be done in various areas.
First of all, only a few research projects have dealt with ciML. The devel-
opment of new ciML approaches for different algorithms has to be expanded to
be able to develop generic human-algorithm interfaces. Research has to focus on
further algorithms beyond decision trees and ant colony algorithms that could
benefit from the new approach of ciML to analyze its full potential.
Secondly, from the knowledge today it cannot be said, which problems ciML
can address and which problems will not be addressable with ciML. Future
research has to focus on the classification of problems in terms of the different
aML, iML and ciML approaches. For some problems we do know that aML can
provide very efficient algorithms, some problems are known to be unsolvable in
polynomial time, but we currently do not have comprehensive knowledge about
the opportunities of ciML.
Thirdly, the iML algorithms proposed so far address very specific problems.
In general, the questions have been solved, how humans can be integrated into
Reasoning Under Uncertainty 371
the algorithm and understand both the underlying problem and the algorithm
with its parameters. Therefore, the past and ongoing research on HCI will play a
prominent role in the future of iML: It has to be further analyzed, how humans
(not only computer scientists) can be empowered to better understand the spe-
cific ML algorithms. This involves adequate visualization techniques of the input
data, as shown by past research projects as well as visualizations to support the
understandability of complex algorithms. In this respect, new interaction tech-
nologies might come in handy. Large displays [76], room-spanning projections
[77], gesture-based interactions and virtual and augmented reality (VR and AR)
[78,79] are new interaction concepts and technologies that have been applied
successfully in the medical [80] and industrial [81–83] domains and might be
able to play a roll in the interaction with algorithms in the future.
6 Conclusion
In this paper, we presented the current state of research in two domains: JDM
and ML. We presented a new classification of ML emphasizing on iML and –
more specificly – on ciML. We bridged the two research domains and argued
that future research will have to take both research domains into account, when
dealing with highly complex problems. Both humans and computers have their
specific strengths and weaknesses and putting humans into the loop of ML algo-
rithms might be a very efficient way for solving specific problems. We identified
two application areas, which provide complex problems that might benefit from
the new approach of ciML: health informatics and cyber-physical systems. While
these two domains seem to be different on the first sight, their problems often
share the same characteristics: Often exceptional variances in data need to be
found, e.g. a specific diseases based on physiological data in medicine or malfunc-
tions of complex cyber-physical systems based on sensor data of machines. The
classical approach of aML focuses on finding these patterns based on previous
knowledge from data. However, aML struggles on function extrapolation prob-
lems which are trivial for human learners. Consequently, integrating a human-
into-the-loop (e.g., a human kernel [84]) could make use of human cognitive
abilities and will be a promising approach. While we outlined the potential of
ciML there are multiple open questions to be tackled in the research community.
The explorative development of new ciML approaches for different algorithms
will help to analyze the full potential of ciML. Existing complex problems need
to be classified and application areas for the different iML approaches need to
be identified. And last but not least, the questions on how to support humans
ideally when collaborating with algorithms and big data needs to be addressed.
In this area the experts from both ML and HCI will have to work hand in hand in
this new joint research endeavor that will greatly help in future problem solving.
Acknowledgements. We thank our colleague Henrik Mucha who provided insight
and expertise that greatly assisted this research. We also thank the anonymous review-
ers for their encouraging reviews.
372 S. Robert et al.
References
1. Holzinger, A.: Trends in interactive knowledge discovery for personalized medicine:
cognitive science meets machine learning. IEEE Intell. Inform. Bull. 15, 6–14 (2014)
2. Von Neumann, J., Morgenstern, O.: Theory of Games and Economic Behavior.
Princeton University Press, Princeton (1944)
3. Fox, J., Glasspool, D., Bury, J.: Quantitative and qualitative approaches to rea-
soning under uncertainty in medical decision making. In: Quaglini, S., Barahona,
P., Andreassen, S. (eds.) AIME 2001. LNCS, vol. 2101, pp. 272–282. Springer,
Heidelberg (2001). doi:10.1007/3-540-48229-6 39
4. Ma, W., Xiong, W., Luo, X.: A model for decision making with missing, imprecise,
and uncertain evaluations of multiple criteria. Int. J. Intell. Syst. 28, 152–184
(2013)
5. Simon, H.A.: A behavioral model of rational choice. Q. J. Econ. 69, 99–118 (1955)
6. Dequech, D.: Bounded rationality, institutions, and uncertainty. J. Econ. Issues
35, 911–929 (2001)
7. Holzinger, A.: Lecture 8 biomedical decision making: reasoning and decision sup-
port. In: Biomedical Informatics, pp. 345–377. Springer, Heidelberg (2014)
8. March, S.T., Hevner, A.R.: Integrated decision support systems: a data warehous-
ing perspective. Decis. Support Syst. 43, 1031–1043 (2007)
9. Hansson, S.O.: Decision theory: a brief introduction (2005)
10. Bell, D.E., Raiffa, H., Tversky, A.: Descriptive, normative, and prescriptive interac-
tions in decision making. Decis. Making Descriptive Normative Prescriptive Inter-
act. 1, 9–32 (1988)
11. Pearl, J.: Heuristics: Intelligent Search Strategies for Computer Problem Solving.
Addison-Wesley, Reading (1984)
12. Alan, D., Janet, F., Gregory, A., Russell, B.: Human-Computer Interaction.
Pearson Education Limited, Harlow (2004)
13. Kohavi, R., Provost, F.: Glossary of terms. Mach. Learn. 30, 271–274 (1998)
14. Ankerst, M., Elsen, C., Ester, M., Kriegel, H.P.: Visual classification: an interactive
approach to decision tree construction. In: Proceedings of the Fifth ACM SIGKDD
International Conference on Knowledge Discovery and Data Mining, pp. 392. ACM
(1999)
15. Wakker, P., Deneffe, D.: Eliciting von neumann-morgenstern utilities when proba-
bilities are distorted or unknown. Manage. Sci. 42, 1131–1150 (1996)
16. Domingos, P.: A few useful things to know about machine learning. Commun.
ACM 55, 78–87 (2012)
17. Mitchell, T.M.: Machine Learning. McGraw-Hill, Boston (1997)
18. Martin, J.H., Jurafsky, D.: Speech and language processing. In: International 710th
edn. (2000)
19. Adomavicius, G., Tuzhilin, A.: Toward the next generation of recommender sys-
tems: a survey of the state-of-the-art and possible extensions. IEEE Trans. Knowl.
Data Eng. 17, 734–749 (2005)
20. Li, Q., Zheng, N., Cheng, H.: Springrobot: a prototype autonomous vehicle and its
algorithms for lane detection. IEEE Trans. Intell. Transp. Syst. 5, 300–308 (2004)
21. Marsland, S.: Machine Learning: An Algorithmic Perspective. CRC Press,
Boca Raton (2015)
22. Ankerst, M., Ester, M., Kriegel, H.P.: Towards an effective cooperation of the user
and the computer for classification. In: Proceedings of the Sixth ACM SIGKDD
International Conference on Knowledge Discovery and Data Mining, pp. 179–188.
ACM (2000)
Reasoning Under Uncertainty 373
23. Ware, M., Frank, E., Holmes, G., Hall, M., Witten, I.H.: Interactive machine learn-
ing: letting users build classifiers. Int. J. Hum. Comput. Stud. 55, 281–292 (2001)
24. Fails, J.A., Olsen Jr., D.R.: Interactive machine learning. In: Proceedings of the
8th International Conference on Intelligent User Interfaces, pp. 39–45. ACM (2003)
25. Fiebrink, R., Cook, P.R., Trueman, D.: Human model evaluation in interactive
supervised learning. In: Proceedings of the SIGCHI Conference on Human Factors
in Computing Systems, CHI 2011, pp. 147–156. ACM, New York (2011)
26. Fogarty, J., Tan, D., Kapoor, A., Winder, S.: Cueflik: interactive concept learning
in image search. In: Proceedings of the SIGCHI Conference on Human Factors in
Computing Systems, CHI 2008, pp. 29–38. ACM, New York (2008)
27. Simard, P., Chickering, D., Lakshmiratan, A., Charles, D., Bottou, L., Suarez,
C.G.J., Grangier, D., Amershi, S., Verwey, J., Suh, J.: Ice: enabling non-experts
to build models interactively for large-scale lopsided problems. arXiv preprint
arXiv:1409.4814 (2014)
28. Amershi, S., Chickering, M., Drucker, S.M., Lee, B., Simard, P., Suh, J.: Mod-
eltracker: redesigning performance analysis tools for machine learning. In: Pro-
ceedings of the 33rd Annual ACM Conference on Human Factors in Computing
Systems, CHI 2015, pp. 337–346. ACM, New York (2015)
29. Talbot, J., Lee, B., Kapoor, A., Tan, D.S.: Ensemblematrix: interactive visual-
ization to support machine learning with multiple classifiers. In: Proceedings of
the SIGCHI Conference on Human Factors in Computing Systems, CHI 2009, pp.
1283–1292. ACM, New York (2009)
30. Ankerst, M., Ester, M., Kriegel, H.P.: Towards an effective cooperation of the user
and the computer for classification. In: Proceedings of the Sixth ACM SIGKDD
International Conference on Knowledge Discovery and Data Mining, KDD 2000,
pp. 179–188. ACM, New York (2000)
31. Holzinger, A.: Interactive machine learning for health informatics: when do we need
the human-in-the-loop? Brain Inform. 3, 119–131 (2016)
32. Holzinger, A., Plass, M., Holzinger, K., Cri¸san, G.C., Pintea, C.-M., Palade, V.:
Towards interactive Machine Learning (iML): applying ant colony algorithms to
solve the traveling salesman problem with the human-in-the-loop approach. In:
Buccafurri, F., Holzinger, A., Kieseberg, P., Tjoa, A.M., Weippl, E. (eds.) CD-
ARES 2016. LNCS, vol. 9817, pp. 81–95. Springer, Heidelberg (2016). doi:10.1007/
978-3-319-45507-5 6
33. Baron, J.: Normative Models of Judgment and Decision Making. Wiley, New York
(2004)
34. Raiffa, H.: Applied statistical decision theory (1974)
35. Murphy, K.P.: Machine Learning: A Probabilistic Perspective. MIT Press,
Cambridge (2012)
36. Friedman, J., Hastie, T., Tibshirani, R.: The Elements of Statistical Learning.
Springer Series in Statistics, vol. 1. Springer, New York (2001)
37. Tulabandhula, T., Rudin, C.: Machine learning with operational costs. J. Mach.
Learn. Res. 14, 1989–2028 (2013)
38. Pitz, G.F., Sachs, N.J.: Judgment and decision: theory and application. Annu. Rev.
Psychol. 35, 139–164 (1984)
39. Fischhoff, B.: Judgment and decision making. Wiley Interdisc. Rev. Cogn. Sci. 1,
724–735 (2010)
40. Russakovsky, O., Li, L.J., Fei-Fei, L.: Best of both worlds: human-machine collabo-
ration for object annotation. In: Proceedings of the IEEE Conference on Computer
Vision and Pattern Recognition, pp. 2121–2131 (2015)
374 S. Robert et al.
41. Rapoport, A.: Decision Theory and Decision Behaviour: Normative and Descriptive
Approaches, vol. 15. Springer, Amsterdam (2013)
42. Bazerman, M.H., Moore, D.A.: Judgment in managerial decision making (2013)
43. Bonner, S.E.: Judgment and Decision Making in Accounting. Prentice Hall,
Upper Saddle River (2008)
44. Robert, S.: Informationsverarbeitung in Prognosen: Experimentelle Evidenz. dis-
sertation, University of Osnabrueck (2016)
45. Goldstein, W.M., Hogarth, R.M.: Research on Judgment and Decision Making:
Currents, Connections, and Controversies. Cambridge University Press, Cambridge
(1997)
46. Milkman, K.L., Chugh, D., Bazerman, M.H.: How can decision making be
improved? Perspect. Psychol. Sci. 4, 379–383 (2009)
47. Baron, J.: Thinking and Deciding. Cambridge University Press, Cambridge (2000)
48. Tversky, A., Kahneman, D.: Judgment under uncertainty: heuristics and biases.
In: Wendt, D., Vlek, C. (eds.) Utility, Probability, and Human Decision Making,
pp. 1124–1131. Springer, Amsterdam (1974)
49. Libby, R.: Accounting and Human Information Processing: Theory and Applica-
tions. Prentice Hall, Englewood Cliffs (1981)
50. Ashton, R.H.: Human Information Processing in Accounting. American Accounting
Association, Sarasota (1982)
51. Over, D.: Rationality and the normative/descriptive distinction. In: Blackwell
Handbook of Judgment and Decision Making, London, pp. 3–18 (2004)
52. Gilovich, T., Griffin, D., Kahneman, D.: Heuristics and Biases: The Psychology of
Intuitive Judgment. Cambridge University Press, New York (2002)
53. Newell, B.R.: Judgment under uncertainty (2013)
54. Tversky, A., Kahneman, D.: Extensional versus intuitive reasoning: the conjunction
fallacy in probability judgment. Psychol. Rev. 90, 293 (1983)
55. Tversky, A., Kahneman, D.: Availability: a heuristic for judging frequency and
probability. Cogn. Psychol. 5, 207–232 (1973)
56. Mokdad, A.H., Marks, J.S., Stroup, D.F., Gerberding, J.L.: Actual causes of death
in the United States, 2000. JAMA 291, 1238–1245 (2004)
57. Strack, F., Mussweiler, T.: Explaining the enigmatic anchoring effect: mechanisms
of selective accessibility. J. Pers. Soc. Psychol. 73, 437 (1997)
58. Plous, S.: Thinking the unthinkable: the effects of anchoring on likelihood estimates
of nuclear war1. J. Appl. Soc. Psychol. 19, 67–91 (1989)
59. Ritov, I.: Anchoring in simulated competitive market negotiation. Organ. Behav.
Hum. Decis. Process. 67, 16–25 (1996)
60. Galinsky, A.D., Mussweiler, T.: First offers as anchors: the role of perspective-
taking and negotiator focus. J. Pers. Soc. Psychol. 81, 657 (2001)
61. Chapman, G.B., Johnson, E.J.: Incorporating the irrelevant: anchors in judgments
of belief and value. In: The Psychology of Intuitive Judgment, Heuristics and
Biases, pp. 120–138 (2002)
62. Wilson, T.D., Houston, C.E., Etling, K.M., Brekke, N.: A new look at anchoring
effects: basic anchoring and its antecedents. J. Exp. Psychol. Gen. 125, 387 (1996)
63. Mussweiler, T., Strack, F.: Comparing is believing: a selective accessibility model
of judgmental anchoring. Eur. Rev. Soc. Psychol. 10, 135–167 (1999)
64. Chapman, G.B., Johnson, E.J.: Anchoring, activation, and the construction of
values. Organ. Behav. Hum. Decis. Process. 79, 115–153 (1999)
65. Furnham, A., Boo, H.C.: A literature review of the anchoring effect. J. Socio-Econ.
40, 35–42 (2011)
Reasoning Under Uncertainty 375
66. Gigerenzer, G.: Why the distinction between single-event probabilities and fre-
quencies is important for psychology (and vice versa). In: Subjective Probability,
pp. 129–161 (1994)
67. Gigerenzer, G., Czerlinski, J., Martignon, L.: How good are fast and frugal heuris-
tics? In: Shanteau, J., Mellers, B.A., Schum, D.A. (eds.) Decision Science and
Technology, pp. 81–103. Springer, New York (1999)
68. Gigerenzer, G., Hoffrage, U., Kleinb¨olting, H.: Probabilistic mental models: a
brunswikian theory of confidence. Psychol. Rev. 98, 506 (1991)
69. Kahneman, D., Tversky, A.: Prospect theory: an analysis of decision under risk.
Econometrica: J. Econometric Soc. 47(2), 263–291 (1979)
70. Xu, L., Jackowski, M., Goshtasby, A., Roseman, D., Bines, S., Yu, C., Dhawan, A.,
Huntley, A.: Segmentation of skin cancer images. Image Vis. Comput. 17, 65–74
(1999)
71. Kr´olczyk, G., Legutko, S., Raos, P.: Cutting wedge wear examination during turn-
ing of duplex stainless steel. Tehniˇcki Vjesnik-Technical Gazette 20, 413–418 (2013)
72. Baron, J.: Rationality and Intelligence. Cambridge University Press, New York
(2005)
73. Lee, S., Holzinger, A.: Knowledge discovery from complex high dimensional data.
In: Michaelis, S., Piatkowski, N., Stolpe, M. (eds.) Solving Large Scale Learning
Tasks. Challenges and Algorithms. LNCS (LNAI), vol. 9580, pp. 148–167. Springer,
Heidelberg (2016). doi:10.1007/978-3-319-41706-6 7
74. Holzinger, A., Malle, B., Giuliani, N.: On graph extraction from image data. In:
Slezak, D., Peters, J.F., Tan, A.H., Schwabe, L. (eds.) Brain Informatics and
Health, BIH 2014. LNAI, vol. 8609, pp. 552–563. Springer, Heidelberg (2014)
75. Valdez, A.C., Dehmer, M., Holzinger, A.: Application of graph entropy
for knowledge discovery and data mining in bibliometric data. In: Dehmer, M.,
Emmert-Streib, F., Chen, Z., Li, X., Shi, Y. (eds.) Mathematical Foundations and
Applications of Graph Entropy, pp. 259–272. Wiley, New York (2016)
76. Cao, X., Balakrishnan, R.: Visionwand: interaction techniques for large displays
using a passive wand tracked in 3d. In: Proceedings of the 16th Annual ACM
Symposium on User Interface Software and Technology, UIST 2003, pp. 173–182.
ACM, New York (2003)
77. Jones, B.R., Benko, H., Ofek, E., Wilson, A.D.: Illumiroom: peripheral projected
illusions for interactive experiences. In: Proceedings of the SIGCHI Conference on
Human Factors in Computing Systems, CHI 2013, pp. 869–878. ACM, New York
(2013)
78. Milgram, P., Takemura, H., Utsumi, A., Kishino, F.: Augmented reality: a class of
displays on the reality-virtuality continuum. In: Photonics for industrial applica-
tions, International Society for Optics and Photonics, pp. 282–292 (1995)
79. Azuma, R.T.: A survey of augmented reality. Presence: Teleoperators Virtual Env-
iron. 6, 355–385 (1997)
80. Fuchs, H., et al.: Augmented reality visualization for laparoscopic surgery. In: Wells,
W.M., Colchester, A., Delp, S. (eds.) MICCAI 1998. LNCS, vol. 1496, pp. 934–943.
Springer, Heidelberg (1998). doi:10.1007/BFb0056282
81. Paelke, V., R¨ocker,C.,Koch,N.,Flatt,H.,B¨uttner, S.: User interfaces for cyber-
physical systems. at-Automatisierungstechnik 63, 833–843 (2015)
82. B¨uttner, S., Sand, O., R¨ocker, C.: Extending the design space in industrial man-
ufacturing through mobile projection. In: Proceedings of the 17th International
Conference on Human-Computer Interaction with Mobile Devices and Services
Adjunct, MobileHCI 2015, pp. 1130–1133. ACM, New York (2015)
376 S. Robert et al.
83. B¨uttner, S., Funk, M., Sand, O., R¨ocker, C.: Using head-mounted displays and
in-situ projection for assistive systems - a comparison. In: Proceedings of the 9th
ACM International Conference on PErvasive Technologies Related to Assistive
Environments, vol. 8. ACM (2016)
84. Wilson, A.G., Dann, C., Lucas, C.G., Xing, E.P.: The human kernel. arXiv preprint
arXiv:1510.07389 (2015)
