Alle Publikationen
- <<
- <
- 1
2019
-
(2019): On Proactive, Transparent, and Verifiable Ethical Reasoning for Robots. In: Proceedings of the IEEE 107 (3), S. 541-561. DOI: 10.1109/JPROC.2019.2898267
DOI: https://doi.org/10.1109/JPROC.2019.2898267 Abstract: Previous work on ethical machine reasoning has largely been theoretical, and where such systems have been implemented, it has, in general, been only initial proofs of principle. Here, we address the question of desirable attributes for such systems to improve their real world utility, and how controllers with these attributes might be implemented. We propose that ethically critical machine reasoning should be proactive, transparent, and verifiable. We describe an architecture where the ethical reasoning is handled by a separate layer, augmenting a typical layered control architecture, ethically moderating the robot actions. It makes use of a simulation-based internal model and supports proactive, transparent, and verifiable ethical reasoning. To do so, the reasoning component of the ethical layer uses our Python-based belief-desire-intention (BDI) implementation. The declarative logic structure of BDI facilitates both transparency, through logging of the reasoning cycle, and formal verification methods. To prove the principles of our approach, we use a case study implementation to experimentally demonstrate its operation. Importantly, it is the first such robot controller where the ethical machine reasoning has been formally verified.
Keywords: BDI implementation, belief desire intention implementation, control engineering computing, Design methodology, ethical machine reasoning, ethical reasoning, Ethics, formal verification, ieee xplore, intelligent robots, layered control architecture, learning (artificial intelligence), machine learning, Moral & Ethik, Predictive models, Python, robot controller, robot programming, Robots, safety, simulation-based internal model, Social implications of technology, software architecture, transparency 2018
-
(2018): A turtle—or a rifle?. Hackers easily fool AIs into seeing the wrong thing. In: Science. DOI: 10.1126/science.aau8383
-
(2018) : ETHICAL FRAMEWORK FOR MACHINE LEARNING: 2018 ITU Kaleidoscope: Machine Learning for a 5G Future (ITU K): Santa Fe, Argentina: IEEE, S. 1-8
DOI: https://doi.org/10.23919/ITU-WT.2018.8597767 Abstract: Artificial Intelligence (AI) with its core subset of Machine Learning (ML) is rapidly transforming life experiences as humans begin to grow more dependent on these ‘smart machines’ for their needs - ranging from routine mundane chores to critical personal decisions. However, these transformative technologies are at the same time proving unpredictable too as has been reported worldwide in certain cases. Therefore, several studies/reports, such as COMEST report on Robotics ethics (UNESCO, 2017) point to an obvious need for inculcating more ethical behavior in machines. The present study aims to look at the role and interplay of ML (the hard sciences) and Ethics (the soft sciences) to resolve such predicaments that are inadvertently manifested by machines not constrained or controlled by human expectations. Based on focused review of literature of both domains-ML and Ethics, the proposed paper attempts to first build on the need for introduction of an ethical algorithm in the domain of machine learning and then endeavors to provide a conceptual framework to resolve the ethical dilemmas.
Keywords: Artificial intelligence, Artificial intelligence/machine learning, Big Data, COMEST report, critical personal decisions, Decision Making, design approach, domains-ML, emotional quotient, ethical algorithm, ethical aspects, ethical behavior, ethical dilemmas, ethical framework, Ethics, hard sciences, human expectations, ieee xplore, learning (artificial intelligence), machine learning, Machine learning algorithms, Moral & Ethik, Prediction algorithms, robotic ethics, Robots, routine mundane chores, smart machines, soft sciences, spiritual quotient, transformative technologies -
(2018): An architecture for ethical robots inspired by the simulation theory of cognition. In: COGNITIVE SYSTEMS RESEARCH 48, S. 56-66. DOI: 10.1016/j.cogsys.2017.04.002
DOI: https://doi.org/10.1016/j.cogsys.2017.04.002 Abstract: The expanding ability of robots to take unsupervised decisions renders it imperative that mechanisms are in place to guarantee the safety of their behaviour. Moreover, intelligent autonomous robots should be more than safe; arguably they should also be explicitly ethical. In this paper, we put forward a method for implementing ethical behaviour in robots inspired by the simulation theory of cognition. In contrast to existing frameworks for robot ethics, our approach does not rely on the verification of logic statements. Rather, it utilises internal simulations which allow the robot to simulate actions and predict their consequences. Therefore, our method is a form of robotic imagery. To demonstrate the proposed architecture, we implement a version of this architecture on a humanoid NAO robot so that it behaves according to Asimov’s laws of robotics. In a series of four experiments, using a second NAO robot as a proxy for the human, we demonstrate that the Ethical Layer enables the robot to prevent the human from coming to harm in simple test scenarios. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
2017
-
(2017): Real‐time gesture–based communication using possibility theory–based hidden Markov model. In: Computational Intelligence 33 (4), S. 843-862. DOI: 10.1111/coin.12116
DOI: https://doi.org/10.1111/coin.12116 Abstract: Exploring correct patterns from low‐frequency time‐series data is challenging. For resolving this problem, the concept of possibility theory–based hidden Markov model (PTBHMM) has been proposed. In this article, all three fundamental problems (evaluation, decoding, and learning) of conventional HMM have been addressed using possibility theory. For handling uncertainty, we have used an axiomatic approach of possibility theory proposed by Zadeh. The time complexity of existing solutions of HMM (forward, backward, Viterbi, and Baum Welch) and proposed possibility‐based solutions has been calculated and compared. From the comparison result, it has been found that PTBHMM has lesser time complexity and hence will be more suitable for real‐time gesture–based communication. (PsycINFO Database Record (c) 2018 APA, all rights reserved)
-
(2017): How Much to Trust Artificial Intelligence?. In: IT Professional 19 (4), S. 7-11. DOI: 10.1109/MITP.2017.3051326
DOI: https://doi.org/10.1109/MITP.2017.3051326 Abstract: Considerable buzz surrounds artificial intelligence, and, indeed, AI is all around us. As with any software-based technology, it is also prone to vulnerabilities. Here, the author examines how we determine whether AI is sufficiently reliable to do its job and how much we should trust its outcomes.
-
(2017) : Acquiring social interaction behaviours for telepresence robots via deep learning from demonstration: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS): Vancouver, British Columbia, Canada: IEEE, S. 37-42
DOI: https://doi.org/10.1109/IROS.2017.8202135 Abstract: As robots begin to inhabit public and social spaces, it is increasingly important to ensure that they behave in a socially appropriate way. However, manually coding social behaviours is prohibitively difficult since social norms are hard to quantify. Therefore, learning from demonstration (LfD), wherein control policies are inferred from demonstrations of correct behaviour, is a powerful tool for helping robots acquire social intelligence. In this paper, we propose a deep learning approach to learning social behaviours from demonstration. We apply this method to two challenging social tasks for a semi-autonomous telepresence robot. Our results show that our approach outperforms gradient boosting regression and performs well against a hard-coded controller. Furthermore, ablation experiments confirm that each element of our method is essential to its success.
Keywords: ablation experiments, challenging social tasks, Cloning, control engineering computing, correct behaviour, deep learning approach, deep learning from demonstration, gradient boosting regression, gradient methods, hard-coded controller, human-robot interaction, ieee xplore, Künstliche Intelligenz, learning (artificial intelligence), LfD, machine learning, public spaces, Regression Analysis, Robot sensing systems, semiautonomous telepresence robot, social behaviour, Social intelligence, social interaction behaviours, Social Norms, social spaces, telepresence robots 2016
-
(2016): Assessing the Moral Coherence and Moral Robustness of Social Systems. Proof of Concept for a Graphical Models Approach. In: Science and engineering ethics 22 (6), S. 1761-1779. DOI: 10.1007/s11948-015-9743-0
DOI: http://www.ncbi.nlm.nih.gov/pubmed/26689413 Abstract: This paper presents a proof of concept for a graphical models approach to assessing the moral coherence and moral robustness of systems of social interactions. "Moral coherence" refers to the degree to which the rights and duties of agents within a system are effectively respected when agents in the system comply with the rights and duties that are recognized as in force for the relevant context of interaction. "Moral robustness" refers to the degree to which a system of social interaction is configured to ensure that the interests of agents are effectively respected even in the face of noncompliance. Using the case of conscientious objection of pharmacists to filling prescriptions for emergency contraception as an example, we illustrate how a graphical models approach can help stakeholders identify structural weaknesses in systems of social interaction and evaluate the relative merits of alternate organizational structures. By illustrating the merits of a graphical models approach we hope to spur further developments in this area.
Keywords: Disziplin, doppelter Treffer, Graphic model concerning "moral robustness" and "moral coherence", graphical models approach, Human Rights, Humans, Kommunikations-/Interaktionsprinzipien, machine learning, Mensch-Technik-Relationen (MTR), Models, Theoretical, Moral, moral agents, moralische Gesetze, moralische Kohärenz, moralische Robustheit, Moralphilosophie, Morals, non/compliance, Philosophie, Respektieren und Erfüllen von Interessen von Agenten, Respektieren von Rechten und Pflichten von Agenten, Social Environment, social interaction, Social Responsibility, Moraltheorie, soziale Interaktion, Soziale Angemessenheit, strukturelle Schwächen in sozialen Systemen, Technikphilosophie, Überblick -
(2016): Socially adaptive path planning in human environments using inverse reinforcement learning. In: International Journal of Social Robotics 8 (1), S. 51-66. DOI: 10.1007/s12369-015-0310-2
DOI: https://doi.org/10.1007/s12369-015-0310-2 Abstract: A key skill for mobile robots is the ability to navigate efficiently through their environment. In the case of social or assistive robots, this involves navigating through human crowds. Typical performance criteria, such as reaching the goal using the shortest path, are not appropriate in such environments, where it is more important for the robot to move in a socially adaptive manner such as respecting comfort zones of the pedestrians. We propose a framework for socially adaptive path planning in dynamic environments, by generating human-like path trajectory. Our framework consists of three modules: a feature extraction module, inverse reinforcement learning (IRL) module, and a path planning module. The feature extraction module extracts features necessary to characterize the state information, such as density and velocity of surrounding obstacles, from a RGB-depth sensor. The inverse reinforcement learning module uses a set of demonstration trajectories generated by an expert to learn the expert’s behaviour when faced with different state features, and represent it as a cost function that respects social variables. Finally, the planning module integrates a three-layer architecture, where a global path is optimized according to a classical shortest-path objective using a global map known a priori, a local path is planned over a shorter distance using the features extracted from a RGB-D sensor and the cost function inferred from IRL module, and a low-level system handles avoidance of immediate obstacles. We evaluate our approach by deploying it on a real robotic wheelchair platform in various scenarios, and comparing the robot trajectories to human trajectories. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
-
(2016): Anticipating Human Activities Using Object Affordances for Reactive Robotic Response. In: IEEE transactions on pattern analysis and machine intelligence 38 (1), S. 14-29. DOI: 10.1109/TPAMI.2015.2430335
DOI: http://www.ncbi.nlm.nih.gov/pubmed/26656575 Abstract: An important aspect of human perception is anticipation, which we use extensively in our day-to-day activities when interacting with other humans as well as with our surroundings. Anticipating which activities will a human do next (and how) can enable an assistive robot to plan ahead for reactive responses. Furthermore, anticipation can even improve the detection accuracy of past activities. The challenge, however, is two-fold: We need to capture the rich context for modeling the activities and object affordances, and we need to anticipate the distribution over a large space of future human activities. In this work, we represent each possible future using an anticipatory temporal conditional random field (ATCRF) that models the rich spatial-temporal relations through object affordances. We then consider each ATCRF as a particle and represent the distribution over the potential futures using a set of particles. In extensive evaluation on CAD-120 human activity RGB-D dataset, we first show that anticipation improves the state-of-the-art detection results. We then show that for new subjects (not seen in the training set), we obtain an activity anticipation accuracy (defined as whether one of top three predictions actually happened) of 84.1, 74.4 and 62.2 percent for an anticipation time of 1, 3 and 10 seconds respectively. Finally, we also show a robot using our algorithm for performing a few reactive responses.
Keywords: Affordance, affordances, Affordanz, Algorithms, Aniticipation, Human Activities, Humans, Imaging, Intentionserkennung (Roboter erkennt Menschenintention), J. J. Gibson's theory of affordance, machine learning, MODELS, Movement, PERCEPTION, Psychological, Robotics, Soziosensitive Systeme, Soziosensitivität, Statistical, Three-Dimensional, Video Recording 2015
-
(2015): A prospective framework for the design of ideal artificial moral agents: Insights from the science of heroism in humans. In: Minds and Machines: Journal for Artificial Intelligence, Philosophy and Cognitive Science 25 (1), S. 57-71. DOI: 10.1007/s11023-015-9361-2
DOI: https://doi.org/10.1007/s11023-015-9361-2 Abstract: The growing field of machine morality has becoming increasingly concerned with how to develop artificial moral agents. However, there is little consensus on what constitutes an ideal moral agent let alone an artificial one. Leveraging a recent account of heroism in humans, the aim of this paper is to provide a prospective framework for conceptualizing, and in turn designing ideal artificial moral agents, namely those that would be considered heroic robots. First, an overview of what it means to be an artificial moral agent is provided. Then, an overview of a recent account of heroism that seeks to define the construct as the dynamic and interactive integration of character strengths (e.g., bravery and integrity) and situational constraints that afford the opportunity for moral behavior (i.e., moral affordances). With this as a foundation, a discussion is provided for what it might mean for a robot to be an ideal moral agent by proposing a dynamic and interactive connectionist model of robotic heroism. Given the limited accounts of robots engaging in moral behavior, a case for extending robotic moral capacities beyond just being a moral agent to the level of heroism is supported by drawing from exemplar situations where robots demonstrate heroism in popular film and fiction. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
2014
-
(2014) : Psychophysiological feedback for adaptive human-robot interaction (HRI) In: Fairclough, Stephen H.; Gilleade, Kiel (Hg.): Advances in physiological computing: New York, NY: Springer-Verlag Publishing (Human-computer interaction series; ISSN: 1571-5035 (Print)), S. 141-167
DOI: https://doi.org/10.1007/978-1-4471-6392-3_7 Abstract: Recent advances in robotics and sensing have given rise to a diverse set of robots and their applications. In recent years robots have increasingly applied in the service industry, search and rescue operations and therapeutic applications. The introduction of robots to interact with humans resulted in a dedicated field called human-robot interaction (HRI). Social HRI is of particular importance as it is the main focus of this chapter. This chapter presents an affect-inspired approach for social HRI. Physiological processing together with machine learning was employed to model affective states for an adaptive social HRI and its application in social interaction in the context of autism therapy was investigated. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
2013
-
(2013): Computationally Modeling Interpersonal Trust. In: Frontiers in psychology 4. DOI: 10.3389/fpsyg.2013.00893
DOI: https://doi.org/10.3389/fpsyg.2013.00893 Abstract: We present a computational model capable of predicting—above human accuracy—the degree of trust a person has toward their novel partner by observing the trust-related nonverbal cues expressed in their social interaction. We summarize our prior work, in which we identify nonverbal cues that signal untrustworthy behavior and also demonstrate the human mind’s readiness to interpret those cues to assess the trustworthiness of a social robot. We demonstrate that domain knowledge gained from our prior work using human-subjects experiments, when incorporated into the feature engineering process, permits a computational model to outperform both human predictions and a baseline model built in naivete' of this domain knowledge. We then present the construction of hidden Markov models to incorporate temporal relationships among the trust-related nonverbal cues. By interpreting the resulting learned structure, we observe that models built to emulate different levels of trust exhibit different sequences of nonverbal cues. From this observation, we derived sequence-based temporal features that further improve the accuracy of our computational model. Our multi-step research process presented in this paper combines the strength of experimental manipulation and machine learning to not only design a computational trust model but also to further our understanding of the dynamics of interpersonal trust.
2011
-
(2011): Problems and issues for service robots in new applications. In: International Journal of Social Robotics 3 (3), S. 299-312. DOI: 10.1007/s12369-011-0097-8
DOI: https://doi.org/10.1007/s12369-011-0097-8 Abstract: Service robots give the possibility of new fields of applications for Robotics by wide spreading robots also into no technical areas. But requirements and goals need to be carefully revised both in designing and operating specific solutions. In this paper, main aspects and challenges are discussed both as problems and advantages in using robots in service operations for new areas of applications, by taking particular attention to non engineering aspects that are deduced from new service areas. In particular, two novel applications, with direct experience of the author and his team, are discussed as referring to robots for restoration activity of historical goods, and robots for physiotherapy rehabilitation and training, as examples with many very different aspects and backgrounds, but even with common issues. Key problems for developing service robots for a successful acceptance and use by even no technical users can be considered in terms of specific technical problems for low-cost user-oriented operation systems, but mainly in terms of implications for human-machine interactions. (PsycINFO Database Record (c) 2016 APA, all rights reserved)
2008
-
(2008) : Learning polite behavior with situation models: 3rd ACM/IEEE International Conference on Human-Robot Interaction (HRI): New York, NY, US: Association for Computing Machinery, S. 209-216
DOI: https://doi.org/10.1145/1349822.1349850 Abstract: In this paper, we describe experiments with methods for learning the appropriateness of behaviors based on a model of the current social situation. We first review different approaches for social robotics, and present a new approach based on situation modeling. We then review algorithms for social learning and propose three modifications to the classical Q-Learning algorithm. We describe five experiments with progressively complex algorithms for learning the appropriateness of behaviors. The first three experiments illustrate how social factors can be used to improve learning by controlling learning rate. In the fourth experiment we demonstrate that proper credit assignment improves the effectiveness of reinforcement learning for social interaction. In our fifth experiment we show that analogy can be used to accelerate learning rates in contexts composed of many situations.
Keywords: Angemessen(heit) (von Technik), Convergence, credit assignment, Humans, ieee xplore, Learning, learning (artificial intelligence), Learning by Analogy, machine learning, polite behavior, Q-Learning, Q-learning algorithm, Reinforcement learning, Robot sensing systems, Robots, situation modeling, social aspects of automation, Social factors, social interaction, social learning, Social robotic, social situation, standards 2007
-
(2007) : The question of ’what to imitate’: Inferring goals and intentions from demonstrations In: Nehaniv, Chrystopher L.; Dautenhahn, Kerstin (Hg.): Imitation and social learning in robots, humans and animals: Behavioural, social and communicative dimensions: New York, NY: Cambridge University Press, S. 135-151
DOI: https://doi.org/10.1017/CBO9780511489808.011 Abstract: A difficult issue for robotics researchers is the question of what to imitate, that is, which aspects of a demonstration robots should copy. Sometimes it is appropriate to copy others’ actions, sometimes it is appropriate to copy others’ results and sometimes copying both or even neither of these is the most appropriate response. The chapter discusses the advantages of using an understanding of others’ goals and intentions to answer this question, copying what the demonstrator intended to do rather than what he actually did. Whereas some animals focus mainly on demonstrators’ results or actions, one-year-old human infants appear to use an understanding of others’ goals to decide what to imitate. The authors identify some specific ways in which infants can infer the goal of a demonstrator in imitation situations, using such information as the demonstrators’ gaze direction, emotional expressions, actions and the context. This is followed by a brief review of what robots currently can do in this regard, proposing some further challenges for them. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
2006
-
(2006) : Learning polite behavior with situation models: 3rd International Forum on Applied Wearable Computing 2006: Bremen, Germany: IEEE, S. 209-216
DOI: https://doi.org/10.1145/1349822.1349850 Abstract: In this paper, we describe experiments with methods for learning the appropriateness of behaviors based on a model of the current social situation. We first review different approaches for social robotics, and present a new approach based on situation modeling. We then review algorithms for social learning and propose three modifications to the classical Q-Learning algorithm. We describe five experiments with progressively complex algorithms for learning the appropriateness of behaviors. The first three experiments illustrate how social factors can be used to improve learning by controlling learning rate. In the fourth experiment we demonstrate that proper credit assignment improves the effectiveness of reinforcement learning for social interaction. In our fifth experiment we show that analogy can be used to accelerate learning rates in contexts composed of many situations.
Keywords: Convergence, credit assignment, Humans, ieee xplore, Künstliche Intelligenz, Learning, learning (artificial intelligence), Learning by Analogy, machine learning, polite behavior, Q-Learning, Q-learning algorithm, Reinforcement learning, Robot sensing systems, Robots, situation modeling, social aspects of automation, Social factors, social interaction, social learning, Social robotic, social situation, standards 2002
-
(2002): Robots that imitate humans. In: Trends in cognitive sciences 6 (11), S. 481-487. DOI: 10.1016/S1364-6613(02)02016-8
DOI: https://doi.org/10.1016/S1364-6613(02)02016-8 Abstract: The study of social learning in robotics has been motivated by both scientific interest in the learning process and practical desires to produce machines that are useful, flexible, and easy to use. In this review, we introduce the social and task-oriented aspects of robot imitation. We focus on methodologies for addressing two fundamental problems. First, how does the robot know what to imitate? And second, how does the robot map that perception onto its own action repertoire to replicate it? In the future, programming humanoid robots to perform new tasks might be as simple as showing them.
-
Dautenhahn, Kerstin; Nehaniv, Chrystopher L. (Hg.) (2002): Imitation in animals and artifacts. Symposium "Imitation in animals and artifacts". Cambridge, Mass: MIT Press ("A Bradford book.")
- <<
- <
- 1