Pub Date : 2017-09-01DOI: 10.1109/COMPANION.2017.8287081
Christian Jarvers, H. Neumann
Companion systems act in and interact with changing environments continuously and in an online manner. Therefore, they are required to adapt to their context of operation in several ways, for example by learning to respond to new input categories. Likewise, reliable tuning of behavior to the current context and to expected future events is necessary. Both types of learning require a trade-off between plasticity (acquiring new concepts or behaviors) and stability (retaining previous knowledge). We outline how dynamic hierarchical networks equipped with a small set of canonical operations can be used to build neural architectures which demonstrate some of the capabilities necessary to fulfill such constraints.
{"title":"Adaptive dynamic network architectures for companion systems","authors":"Christian Jarvers, H. Neumann","doi":"10.1109/COMPANION.2017.8287081","DOIUrl":"https://doi.org/10.1109/COMPANION.2017.8287081","url":null,"abstract":"Companion systems act in and interact with changing environments continuously and in an online manner. Therefore, they are required to adapt to their context of operation in several ways, for example by learning to respond to new input categories. Likewise, reliable tuning of behavior to the current context and to expected future events is necessary. Both types of learning require a trade-off between plasticity (acquiring new concepts or behaviors) and stability (retaining previous knowledge). We outline how dynamic hierarchical networks equipped with a small set of canonical operations can be used to build neural architectures which demonstrate some of the capabilities necessary to fulfill such constraints.","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116855125","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/COMPANION.2017.8287080
Siwen Guo, Christoph Schommer
The term Artificial Companion has originally been introduced by Y. Wilks [1] as “…an intelligent and helpful cognitive agent, which appears to know its owner and their habits, chats to them and diverts them, assists them with simple tasks…”. To serve the users' interests by considering a personal knowledge is, furthermore, demanded. The following position paper takes this request as motivation for the embedding of the PERSEUS1 system, which is a personalized sentiment framework based on a Deep Learning approach. We discuss how such an embedding with a group of users should be realized and why the utilization of PERSEUS is beneficial.
{"title":"Embedding of the personalized sentiment engine PERSEUS in an artificial companion","authors":"Siwen Guo, Christoph Schommer","doi":"10.1109/COMPANION.2017.8287080","DOIUrl":"https://doi.org/10.1109/COMPANION.2017.8287080","url":null,"abstract":"The term Artificial Companion has originally been introduced by Y. Wilks [1] as “…an intelligent and helpful cognitive agent, which appears to know its owner and their habits, chats to them and diverts them, assists them with simple tasks…”. To serve the users' interests by considering a personal knowledge is, furthermore, demanded. The following position paper takes this request as motivation for the embedding of the PERSEUS1 system, which is a personalized sentiment framework based on a Deep Learning approach. We discuss how such an embedding with a group of users should be realized and why the utilization of PERSEUS is beneficial.","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124100115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/COMPANION.2017.8287079
Olga Egorow, A. Lotz, Ingo Siegert, Ronald Böck, J. Krüger, A. Wendemuth
One objective of affective computing is the automatic processing of human emotions. Considering human speech, filled pauses are one of the cues giving insight into the emotional state of a human being. Filled pauses are short speech events without a specified semantic meaning, but they have a variety of communicative and affective functions. The detection and processing of such speech events can help a technical system to recognise the affective state of the user. To solve this task using machine learning methods, huge amounts of annotated data and thus human resources are necessary. In this paper we introduce an efficient approach for semiautomatic labelling of filled pauses aiming at finding as many of them as possible with minimal effort. We investigate to which extent such an approach can reduce the effort of manual transcription of filled pauses. By using our approach, we could for the first time quantify that the time necessary for the human supervised verification can be reduced by up to 85% compared to a full manual annotation.
{"title":"Accelerating manual annotation of filled pauses by automatic pre-selection","authors":"Olga Egorow, A. Lotz, Ingo Siegert, Ronald Böck, J. Krüger, A. Wendemuth","doi":"10.1109/COMPANION.2017.8287079","DOIUrl":"https://doi.org/10.1109/COMPANION.2017.8287079","url":null,"abstract":"One objective of affective computing is the automatic processing of human emotions. Considering human speech, filled pauses are one of the cues giving insight into the emotional state of a human being. Filled pauses are short speech events without a specified semantic meaning, but they have a variety of communicative and affective functions. The detection and processing of such speech events can help a technical system to recognise the affective state of the user. To solve this task using machine learning methods, huge amounts of annotated data and thus human resources are necessary. In this paper we introduce an efficient approach for semiautomatic labelling of filled pauses aiming at finding as many of them as possible with minimal effort. We investigate to which extent such an approach can reduce the effort of manual transcription of filled pauses. By using our approach, we could for the first time quantify that the time necessary for the human supervised verification can be reduced by up to 85% compared to a full manual annotation.","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"66 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114050879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/COMPANION.2017.8287077
G. Behnke, F. Nielsen, Marvin R. G. Schiller, P. Bercher, Matthias Kraus, W. Minker, Birte Glimm, Susanne Biundo-Stephan
We present the mixed-initiative planning system Sloth, which is designed to assist users in planning a fitness workout. Mixed-initiative planning systems are especially useful for companion systems, as they allow the seamless integration of the complex cognitive ability of planning into ambient assistance systems. This is achieved by integrating the user directly into the process of plan generation and thereby allowing him to specify these objectives and to be assisted in generating a plan that not only achieves his objectives, but at the same time also fits his preferences. We present the capabilities that are integrated into Sloth and discuss the design choices and considerata that have to be taken into account when constructing a mixed-initiative planning system.
{"title":"Sloth — The interactive workout planner","authors":"G. Behnke, F. Nielsen, Marvin R. G. Schiller, P. Bercher, Matthias Kraus, W. Minker, Birte Glimm, Susanne Biundo-Stephan","doi":"10.1109/COMPANION.2017.8287077","DOIUrl":"https://doi.org/10.1109/COMPANION.2017.8287077","url":null,"abstract":"We present the mixed-initiative planning system Sloth, which is designed to assist users in planning a fitness workout. Mixed-initiative planning systems are especially useful for companion systems, as they allow the seamless integration of the complex cognitive ability of planning into ambient assistance systems. This is achieved by integrating the user directly into the process of plan generation and thereby allowing him to specify these objectives and to be assisted in generating a plan that not only achieves his objectives, but at the same time also fits his preferences. We present the capabilities that are integrated into Sloth and discuss the design choices and considerata that have to be taken into account when constructing a mixed-initiative planning system.","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122802447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ICCT42709.2017.9151907
G. Behnke, B. Leichtmann, P. Bercher, D. Höller, V. Nitsch, M. Baumann, Susanne Biundo-Stephan
A promising field of application for cognitive technical systems is individualised user assistance for complex tasks. Here, a companion system usually uses an AI planner to solve the underlying combinatorial problem. Often, the use of a bare black-box planning system is not sufficient to provide individualised assistance, but instead the user has to be able to control the process that generates the presented advice. Such an integration guarantees that the user will be satisfied with the assistance s/he is given, trust the advice more, and is thus more likely to follow it. In this paper, we provide a general theoretical view on this process, called mixed-initiative planning, and derive several research challenges from it.
{"title":"Help me make a dinner! Challenges when assisting humans in action planning","authors":"G. Behnke, B. Leichtmann, P. Bercher, D. Höller, V. Nitsch, M. Baumann, Susanne Biundo-Stephan","doi":"10.1109/ICCT42709.2017.9151907","DOIUrl":"https://doi.org/10.1109/ICCT42709.2017.9151907","url":null,"abstract":"A promising field of application for cognitive technical systems is individualised user assistance for complex tasks. Here, a companion system usually uses an AI planner to solve the underlying combinatorial problem. Often, the use of a bare black-box planning system is not sufficient to provide individualised assistance, but instead the user has to be able to control the process that generates the presented advice. Such an integration guarantees that the user will be satisfied with the assistance s/he is given, trust the advice more, and is thus more likely to follow it. In this paper, we provide a general theoretical view on this process, called mixed-initiative planning, and derive several research challenges from it.","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126462422","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/COMPANION.2017.8287082
Kristiina Jokinen, Satoshi Nishimura, Ken Fukuda, Takuichi Nishimura
The paper focuses on issues related to dialogue modelling that enables interaction between users and service systems, on topics that concern the experience and knowledge of people in the service industry. We discuss dialogue design that is based on the crowd-sourced data structured according to goal-directed ontological requirements, as well as architectural aspects and the human-centered companion view of IoT communication. Our ultimate aim is to create a framework for improving and reconstructing operations for service industries such as nursing, caregiving, and education, and for activities that promote health and create community through hobbies such as dance and music.
{"title":"Dialogues with IoT companions: Enabling human interaction with intelligent service items","authors":"Kristiina Jokinen, Satoshi Nishimura, Ken Fukuda, Takuichi Nishimura","doi":"10.1109/COMPANION.2017.8287082","DOIUrl":"https://doi.org/10.1109/COMPANION.2017.8287082","url":null,"abstract":"The paper focuses on issues related to dialogue modelling that enables interaction between users and service systems, on topics that concern the experience and knowledge of people in the service industry. We discuss dialogue design that is based on the crowd-sourced data structured according to goal-directed ontological requirements, as well as architectural aspects and the human-centered companion view of IoT communication. Our ultimate aim is to create a framework for improving and reconstructing operations for service industries such as nursing, caregiving, and education, and for activities that promote health and create community through hobbies such as dance and music.","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"313 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114414561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/COMPANION.2017.8287084
Andreas Oschlies-Strobel, Sascha Gruss, L. Jerg-Bretzke, Steffen Walter, Dilana Hazer-Rau
In this work, different cognitive load situations are examined and classified in the context of a Human Computer Interaction (HCI) scenario. Machine learning methods were used to detect three cognitive load states (overload, underload, normal load) with the help of five different psychophysiological signals (ECG, EMG, Respiration, GSR, Temperature). At first it is shown, that the three regarded states can be clearly distinguished in the Valence-Arousal-Dominance space (VAD). After this comparisons between a 10-fold-valdidation and a batch-validation as well as three different classifiers (k-Nearest-Neighbour, Naive Bayes, Random Forest) are accomplished. At last the influence of gender in contrast to an overall analysis is shown.
{"title":"Preliminary classification of cognitive load states in a human machine interaction scenario","authors":"Andreas Oschlies-Strobel, Sascha Gruss, L. Jerg-Bretzke, Steffen Walter, Dilana Hazer-Rau","doi":"10.1109/COMPANION.2017.8287084","DOIUrl":"https://doi.org/10.1109/COMPANION.2017.8287084","url":null,"abstract":"In this work, different cognitive load situations are examined and classified in the context of a Human Computer Interaction (HCI) scenario. Machine learning methods were used to detect three cognitive load states (overload, underload, normal load) with the help of five different psychophysiological signals (ECG, EMG, Respiration, GSR, Temperature). At first it is shown, that the three regarded states can be clearly distinguished in the Valence-Arousal-Dominance space (VAD). After this comparisons between a 10-fold-valdidation and a batch-validation as well as three different classifiers (k-Nearest-Neighbour, Naive Bayes, Random Forest) are accomplished. At last the influence of gender in contrast to an overall analysis is shown.","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131252519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/COMPANION.2017.8287076
M. Siebers, Kyra Gobel, C. Niessen, Ute Schmid
Though forgetting seems to be an everyday hassle, forgetting information not needed in the current context is helpful, even necessary. Such unneeded information is irrelevant to your current task at best, and at worst, deteriorates performance. For example, when writing a computer program in C#, you should temporarily forget your Java knowledge, since these programming languages are similar but not quite the same. Bjork, Bjork, and Anderson summarize this type of forgetting under the term goal-directed forgetting [1].
{"title":"Requirements for a companion system to support identifying irrelevancy","authors":"M. Siebers, Kyra Gobel, C. Niessen, Ute Schmid","doi":"10.1109/COMPANION.2017.8287076","DOIUrl":"https://doi.org/10.1109/COMPANION.2017.8287076","url":null,"abstract":"Though forgetting seems to be an everyday hassle, forgetting information not needed in the current context is helpful, even necessary. Such unneeded information is irrelevant to your current task at best, and at worst, deteriorates performance. For example, when writing a computer program in C#, you should temporarily forget your Java knowledge, since these programming languages are similar but not quite the same. Bjork, Bjork, and Anderson summarize this type of forgetting under the term goal-directed forgetting [1].","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131476043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/COMPANION.2017.8287083
Viktor Kessler, Patrick Thiam, Mohammadreza Amirian, F. Schwenker
The research in classifying affective states of a participant provided a great amount of feature extraction methods in several modalities like facial motion, speech, biophysiological signals and Action Units (AU). The ability of predicting the heart rate of a participant with remote Photoplethysmography (rPPG) from the video channel enables an interesting modality for classification of affective states but only few authors tried it. In this work, we present the rPPG signal as a new modality for pain classification and evaluate the benefit of a fusion with other modalities. In short the rPPG signal is filtered in multiple frequency ranges corresponding to the respiration rate as biophysiological signal. Then the pain is classified by fusing all modalities with a hierarchical fusion architecture. The performance could be increased around ∼1.4% with the rPPG signal even in combination with biophysiological signals from a biosignal amplifier.
{"title":"Multimodal fusion including camera photoplethysmography for pain recognition","authors":"Viktor Kessler, Patrick Thiam, Mohammadreza Amirian, F. Schwenker","doi":"10.1109/COMPANION.2017.8287083","DOIUrl":"https://doi.org/10.1109/COMPANION.2017.8287083","url":null,"abstract":"The research in classifying affective states of a participant provided a great amount of feature extraction methods in several modalities like facial motion, speech, biophysiological signals and Action Units (AU). The ability of predicting the heart rate of a participant with remote Photoplethysmography (rPPG) from the video channel enables an interesting modality for classification of affective states but only few authors tried it. In this work, we present the rPPG signal as a new modality for pain classification and evaluate the benefit of a fusion with other modalities. In short the rPPG signal is filtered in multiple frequency ranges corresponding to the respiration rate as biophysiological signal. Then the pain is classified by fusing all modalities with a hierarchical fusion architecture. The performance could be increased around ∼1.4% with the rPPG signal even in combination with biophysiological signals from a biosignal amplifier.","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"19 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123448717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/COMPANION.2017.8287073
Thorsten Schodde, Laura Hoffmann, S. Kopp
Social robots represent a fruitful enhancement of intelligent tutoring systems that can be used for one-to-one tutoring. The role of affective states during learning has so far only scarcely been considered in such systems, because it is unclear which cues should be tracked, how they should be interpreted, and how the system should react to them. Therefore, we conducted expert interviews with preschool teachers, and based on these results suggest a conceptual model for tracing and managing the affective state of preschool children during robot-child tutoring.
{"title":"How to manage affective state in child-robot tutoring interactions?","authors":"Thorsten Schodde, Laura Hoffmann, S. Kopp","doi":"10.1109/COMPANION.2017.8287073","DOIUrl":"https://doi.org/10.1109/COMPANION.2017.8287073","url":null,"abstract":"Social robots represent a fruitful enhancement of intelligent tutoring systems that can be used for one-to-one tutoring. The role of affective states during learning has so far only scarcely been considered in such systems, because it is unclear which cues should be tracked, how they should be interpreted, and how the system should react to them. Therefore, we conducted expert interviews with preschool teachers, and based on these results suggest a conceptual model for tracing and managing the affective state of preschool children during robot-child tutoring.","PeriodicalId":132735,"journal":{"name":"2017 International Conference on Companion Technology (ICCT)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124612455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：