Pub Date : 2013-12-31DOI: 10.1109/DEVLRN.2013.6652561
J. Lones, L. Cañamero
Epigenetic adaptation provides biological organisms with the ability to adjust their physiology and/or morphology in order to meet some of the challenges posed by their environment. Recent research has suggested that this process may be controlled by hormones. In this paper, we present a model that allows an autonomous robot to develop its systems in accordance with the environment it is currently situated in. Experiments have been undertaken in multiple environments with different challenges and niches to negotiate. We have so far seen encouraging results and the emergence of unique behaviours tailored to exploiting its current environment.
{"title":"Epigenetic adaptation through hormone modulation in autonomous robots","authors":"J. Lones, L. Cañamero","doi":"10.1109/DEVLRN.2013.6652561","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652561","url":null,"abstract":"Epigenetic adaptation provides biological organisms with the ability to adjust their physiology and/or morphology in order to meet some of the challenges posed by their environment. Recent research has suggested that this process may be controlled by hormones. In this paper, we present a model that allows an autonomous robot to develop its systems in accordance with the environment it is currently situated in. Experiments have been undertaken in multiple environments with different challenges and niches to negotiate. We have so far seen encouraging results and the emergence of unique behaviours tailored to exploiting its current environment.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123872979","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 : 2013-11-04DOI: 10.1109/DEVLRN.2013.6652555
S. Bambach, David J. Crandall, Chen Yu
A key component of the human visual system is our attentional control - the selection of which visual stimuli to pay attention to at any moment in time. Understanding visual attention in children could yield new insight into how the visual system develops during formative years and how their visual attention and selection play a role in development and learning. We use head-mounted cameras to record first-person video from interacting children and parents, giving a good approximation of the contents of their visual fields of view, and collect gaze direction data to record where they look within the visual field. We data-mine this data to study the distributions of gaze patterns within the first-person visual frame for both children and adults. We also study the ability of visual saliency to predict visual attention, as a function of the tasks, actions, and interactions that the participants perform. We find significant differences in the results between children and parents, indicating substantial differences in how their bodily actions are coupled with their visual attention between developing (child) and developed (adult) visual systems.
{"title":"Understanding embodied visual attention in child-parent interaction","authors":"S. Bambach, David J. Crandall, Chen Yu","doi":"10.1109/DEVLRN.2013.6652555","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652555","url":null,"abstract":"A key component of the human visual system is our attentional control - the selection of which visual stimuli to pay attention to at any moment in time. Understanding visual attention in children could yield new insight into how the visual system develops during formative years and how their visual attention and selection play a role in development and learning. We use head-mounted cameras to record first-person video from interacting children and parents, giving a good approximation of the contents of their visual fields of view, and collect gaze direction data to record where they look within the visual field. We data-mine this data to study the distributions of gaze patterns within the first-person visual frame for both children and adults. We also study the ability of visual saliency to predict visual attention, as a function of the tasks, actions, and interactions that the participants perform. We find significant differences in the results between children and parents, indicating substantial differences in how their bodily actions are coupled with their visual attention between developing (child) and developed (adult) visual systems.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134430615","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 : 2013-11-04DOI: 10.1109/DEVLRN.2013.6652568
S. Doncieux
In the perspective of life long learning, a robot may face different, but related situations. Being able to exploit the knowledge acquired during a first learning phase may be critical in order to solve more complex tasks. This is the transfer learning problem. This problem is addressed here in the case of direct policy search algorithms. No discrete states, nor actions are defined a priori. A policy is described by a controller that computes orders to be sent to the motors out of sensor values. Both motor and sensor values can be continuous. The proposed approach relies on population based direct policy search algorithms, i.e. evolutionary algorithms. It exploits the numerous behaviors that are generated during the search. When learning on the source task, a knowledge base is built. The knowledge base aims at identifying the most salient behaviors segments with regards to the considered task. Afterwards, the knowledge base is exploited on a target task, with a reward shaping approach: besides its reward on the task, a policy is credited with a reward computed from the knowledge base. The rationale behind this approach is to automatically detect the stepping stones, i.e. the behavior segments that have lead to a reward in the source task before the policy is efficient enough to get the reward on the target task. The approach is tested in simulation with a neuroevolution approach and on ball collecting tasks.
{"title":"Transfer learning for direct policy search: A reward shaping approach","authors":"S. Doncieux","doi":"10.1109/DEVLRN.2013.6652568","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652568","url":null,"abstract":"In the perspective of life long learning, a robot may face different, but related situations. Being able to exploit the knowledge acquired during a first learning phase may be critical in order to solve more complex tasks. This is the transfer learning problem. This problem is addressed here in the case of direct policy search algorithms. No discrete states, nor actions are defined a priori. A policy is described by a controller that computes orders to be sent to the motors out of sensor values. Both motor and sensor values can be continuous. The proposed approach relies on population based direct policy search algorithms, i.e. evolutionary algorithms. It exploits the numerous behaviors that are generated during the search. When learning on the source task, a knowledge base is built. The knowledge base aims at identifying the most salient behaviors segments with regards to the considered task. Afterwards, the knowledge base is exploited on a target task, with a reward shaping approach: besides its reward on the task, a policy is credited with a reward computed from the knowledge base. The rationale behind this approach is to automatically detect the stepping stones, i.e. the behavior segments that have lead to a reward in the source task before the policy is efficient enough to get the reward on the target task. The approach is tested in simulation with a neuroevolution approach and on ball collecting tasks.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"6 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132336138","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 : 2013-11-04DOI: 10.1109/DEVLRN.2013.6652542
Sumarga H. Suanda, Seth B. Foster, Linda B. Smith, Chen Yu
Recent research supports the notion that word learning can be conceptualized as a statistical learning process. As many have noted however, statistical learning is constrained by processes such as attention and memory. In the current study, we observed, through toddler-perspective head cameras, toddlers' visual input as parents labeled novel objects during an object-play session. We then analyzed the co-occurrence statistics between words and objects that accumulated over the session. We also analyzed the constrained co-occurrence statistics which took into consideration the perceptual properties of the objects (e.g., object size) at the times words were uttered. We compared the information in these two types of statistical structures and examined which of the two best fit with the patterns of children's object-name learning. Implications of these results for statistical learning accounts of early word learning are discussed.
{"title":"Attentional constraints and statistics in toddlers' word learning","authors":"Sumarga H. Suanda, Seth B. Foster, Linda B. Smith, Chen Yu","doi":"10.1109/DEVLRN.2013.6652542","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652542","url":null,"abstract":"Recent research supports the notion that word learning can be conceptualized as a statistical learning process. As many have noted however, statistical learning is constrained by processes such as attention and memory. In the current study, we observed, through toddler-perspective head cameras, toddlers' visual input as parents labeled novel objects during an object-play session. We then analyzed the co-occurrence statistics between words and objects that accumulated over the session. We also analyzed the constrained co-occurrence statistics which took into consideration the perceptual properties of the objects (e.g., object size) at the times words were uttered. We compared the information in these two types of statistical structures and examined which of the two best fit with the patterns of children's object-name learning. Implications of these results for statistical learning accounts of early word learning are discussed.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114611348","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 : 2013-11-04DOI: 10.1109/DEVLRN.2013.6652543
Francesca Stramandinoli, D. Marocco, A. Cangelosi
Cognitive developmental robotics is facing the challenge of building robots capable of working independently and/or with other agents in non-structured scenarios, which can autonomously react to dynamic changes that occur in the environment. Providing robots with the capability to comprehend and produce language in a “human-like” manner represents a powerful tool for flexible and intelligent interaction between robots and human beings. Robots endowed with linguistic capabilities, in fact, could better understand situations and exchange information; through language robots could cooperate and negotiate with human beings in order to accomplish shared plans. This work describes a neuro-robotics model for the acquisition of abstract action words in the iCub humanoid robot. We claim that the acquisition of concepts that refer to such abstract words (e.g. verbs like “use”, “make”) can be driven by action's organization. In the presented model the integration of low-level capabilities (e.g. perceptual and sensorimotor skills) enables the hierarchical organization of concepts that leads to the grounding of more general words.
{"title":"Grounding abstract action words through the hierarchical organization of motor primitives","authors":"Francesca Stramandinoli, D. Marocco, A. Cangelosi","doi":"10.1109/DEVLRN.2013.6652543","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652543","url":null,"abstract":"Cognitive developmental robotics is facing the challenge of building robots capable of working independently and/or with other agents in non-structured scenarios, which can autonomously react to dynamic changes that occur in the environment. Providing robots with the capability to comprehend and produce language in a “human-like” manner represents a powerful tool for flexible and intelligent interaction between robots and human beings. Robots endowed with linguistic capabilities, in fact, could better understand situations and exchange information; through language robots could cooperate and negotiate with human beings in order to accomplish shared plans. This work describes a neuro-robotics model for the acquisition of abstract action words in the iCub humanoid robot. We claim that the acquisition of concepts that refer to such abstract words (e.g. verbs like “use”, “make”) can be driven by action's organization. In the presented model the integration of low-level capabilities (e.g. perceptual and sensorimotor skills) enables the hierarchical organization of concepts that leads to the grounding of more general words.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"49 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128529290","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 : 2013-11-04DOI: 10.1109/DEVLRN.2013.6652553
D. Takahashi, Darshana Narayanan, A. Ghazanfar
In humans, self-monitoring is essential for conversations [1] and this ability is fully developed only after 2 years of age [2]. How vocal self-monitoring changes over the course of development, how it evolved, and how it is mediated by neural mechanisms are not known. In this study, we use marmoset monkeys to address these issues. We recorded natural vocal exchanges between parent and infant monkeys and looked for evidence of self-monitoring in both the adults and the infants. Our results show that, similar to humans, adult marmosets have the capacity to self-monitor whereas infants do not. Using a computational model of marmoset vocal exchange dynamics [3] we investigated possible neural mechanism underlying the development of self-monitoring. Our model is based on the interactions between three neural structures (representing limbic, motor and auditory regions) with feedback connectivity inspired by published physiological and anatomical data. The simulations show that strengthening the inhibitory connection between auditory and limbic regions drives the development of self-monitoring and, ultimately, adult-like vocal behavior.
{"title":"Development of self-monitoring essential for vocal interactions in marmoset monkeys","authors":"D. Takahashi, Darshana Narayanan, A. Ghazanfar","doi":"10.1109/DEVLRN.2013.6652553","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652553","url":null,"abstract":"In humans, self-monitoring is essential for conversations [1] and this ability is fully developed only after 2 years of age [2]. How vocal self-monitoring changes over the course of development, how it evolved, and how it is mediated by neural mechanisms are not known. In this study, we use marmoset monkeys to address these issues. We recorded natural vocal exchanges between parent and infant monkeys and looked for evidence of self-monitoring in both the adults and the infants. Our results show that, similar to humans, adult marmosets have the capacity to self-monitor whereas infants do not. Using a computational model of marmoset vocal exchange dynamics [3] we investigated possible neural mechanism underlying the development of self-monitoring. Our model is based on the interactions between three neural structures (representing limbic, motor and auditory regions) with feedback connectivity inspired by published physiological and anatomical data. The simulations show that strengthening the inhibitory connection between auditory and limbic regions drives the development of self-monitoring and, ultimately, adult-like vocal behavior.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130958789","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 : 2013-11-04DOI: 10.1109/DEVLRN.2013.6652548
Yasunori Yamada, Keiko Fujii, Y. Kuniyoshi
Recent developmental studies have shown the importance of interaction with the environment resulting from fetal spontaneous movements for the formation of body maps in the spinal cord and primary somatosensory area. However, the underlying mechanism as well as the factors which contribute to the development of body maps is largely unknown. Here, we simulated the development of the body map using a human fetus simulation, and investigated the contribution of three factors which often differ in normal fetuses and preterms: (i) developmental environment, (ii) nervous system and (iii) movement patterns. The fetus model has a musculoskeletal body as well as sensory organs for tactile and proprioception, which allows us to simulate sensory feedbacks resulting from interaction with the uterine environment. We simulated the development of body maps by using these sensory feedbacks as inputs to the spinal and somatosensory cortex models, which are spiking neural networks with leaky integrate-and-fire neurons and spike-timing-dependent synaptic plasticity. We showed that the networks under normal fetus conditions can learn body part-specific modular architectures and responses, and have neurons encoding specific postures and double-touch events. In contrast, we found that (i) a change in environment from inside to outside the uterus, (ii) an imbalance between excitation and inhibition of the nervous system and (iii) decrease in variation and complexity of movement patterns each lead to the development of abnormal body maps in terms of function and structure of learned networks. These results suggest that these factors influence or disrupt body map development in preterms and their subsequent cognitive development.
{"title":"Impacts of environment, nervous system and movements of preterms on body map development: Fetus simulation with spiking neural network","authors":"Yasunori Yamada, Keiko Fujii, Y. Kuniyoshi","doi":"10.1109/DEVLRN.2013.6652548","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652548","url":null,"abstract":"Recent developmental studies have shown the importance of interaction with the environment resulting from fetal spontaneous movements for the formation of body maps in the spinal cord and primary somatosensory area. However, the underlying mechanism as well as the factors which contribute to the development of body maps is largely unknown. Here, we simulated the development of the body map using a human fetus simulation, and investigated the contribution of three factors which often differ in normal fetuses and preterms: (i) developmental environment, (ii) nervous system and (iii) movement patterns. The fetus model has a musculoskeletal body as well as sensory organs for tactile and proprioception, which allows us to simulate sensory feedbacks resulting from interaction with the uterine environment. We simulated the development of body maps by using these sensory feedbacks as inputs to the spinal and somatosensory cortex models, which are spiking neural networks with leaky integrate-and-fire neurons and spike-timing-dependent synaptic plasticity. We showed that the networks under normal fetus conditions can learn body part-specific modular architectures and responses, and have neurons encoding specific postures and double-touch events. In contrast, we found that (i) a change in environment from inside to outside the uterus, (ii) an imbalance between excitation and inhibition of the nervous system and (iii) decrease in variation and complexity of movement patterns each lead to the development of abnormal body maps in terms of function and structure of learned networks. These results suggest that these factors influence or disrupt body map development in preterms and their subsequent cognitive development.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134243443","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 : 2013-11-04DOI: 10.1109/DEVLRN.2013.6652573
Bradley Hayes, B. Scassellati
One of the hallmarks of development is the transition of an agent from novice learner to able partner to experienced instructor. While most machine learning approaches focus on the first transition, we are interested in building an effective learning and development system that allows for the complete range of transitions to occur. In this paper, we present a mechanism enabling such transitions within the context of collaborative social tasks. We present a cooperative robot system capable of learning a hierarchical task execution from an experienced human user, collaborating safely with a knowledgeable human peer, and instructing a novice user based on the explicit inclusion of a feature within the planning and skill execution subsystems we've termed social force. We conclude with an evaluation of this feature's flexibility within a collaborative construction task, changing a robot's behaviors between student, peer, and instructor through simple manipulations of this feature's treatment within the planning subsystem.
{"title":"Improving implicit communication in mixed human-robot teams with social force detection","authors":"Bradley Hayes, B. Scassellati","doi":"10.1109/DEVLRN.2013.6652573","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652573","url":null,"abstract":"One of the hallmarks of development is the transition of an agent from novice learner to able partner to experienced instructor. While most machine learning approaches focus on the first transition, we are interested in building an effective learning and development system that allows for the complete range of transitions to occur. In this paper, we present a mechanism enabling such transitions within the context of collaborative social tasks. We present a cooperative robot system capable of learning a hierarchical task execution from an experienced human user, collaborating safely with a knowledgeable human peer, and instructing a novice user based on the explicit inclusion of a feature within the planning and skill execution subsystems we've termed social force. We conclude with an evaluation of this feature's flexibility within a collaborative construction task, changing a robot's behaviors between student, peer, and instructor through simple manipulations of this feature's treatment within the planning subsystem.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"327 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124297165","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 : 2013-11-04DOI: 10.1109/DEVLRN.2013.6652526
Matthias Rolf
Goal babbling is a recent concept for the efficient bootstrapping of sensorimotor coordination that is inspired by infants' early goal-directed movement attempts. Several studies have shown its superior performance compared to random motor babbling. Yet, previous implementations of goal babbling require knowledge of a set of achievable goals in advance. This paper introduces an approach to goal babbling that can bootstrap coordination skills without pre-specifying, or even representing, a set of goals. On the contrary, it can discover the ranges of achievable goals autonomously. This capability is demonstrated in a challenging task with up to 50 degrees of freedom, in which the discovery of possible outcomes is shown to be desperately intractable with random motor babbling.
{"title":"Goal babbling with unknown ranges: A direction-sampling approach","authors":"Matthias Rolf","doi":"10.1109/DEVLRN.2013.6652526","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652526","url":null,"abstract":"Goal babbling is a recent concept for the efficient bootstrapping of sensorimotor coordination that is inspired by infants' early goal-directed movement attempts. Several studies have shown its superior performance compared to random motor babbling. Yet, previous implementations of goal babbling require knowledge of a set of achievable goals in advance. This paper introduces an approach to goal babbling that can bootstrap coordination skills without pre-specifying, or even representing, a set of goals. On the contrary, it can discover the ranges of achievable goals autonomously. This capability is demonstrated in a challenging task with up to 50 degrees of freedom, in which the discovery of possible outcomes is shown to be desperately intractable with random motor babbling.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"692 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126341665","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 : 2013-11-04DOI: 10.1109/DEVLRN.2013.6652554
A. Karaouzene, P. Gaussier, Denis Vidal
In this work, we present a model based on social referencing mechanisms for the artwork appreciation development. We show that trying to reach autonomous robot artwork preferences is an interesting framework for developmental robotics. Therefore, we present an application of our model in a natural environment. The museum as an experiment place, helps us to benefit from a large number of visitors with different backgrounds and personal interests, as well as provides the challenge of the learning and recognition of an important number of artefacts (heterogeneous set of objects in terms of size, culture). To overcome some limitations like the autonomy in the learning, we proposed a first model for cumulative learning using a second-order conditioning approach with very encouraging results.
{"title":"A robot to study the development of artwork appreciation through social interactions","authors":"A. Karaouzene, P. Gaussier, Denis Vidal","doi":"10.1109/DEVLRN.2013.6652554","DOIUrl":"https://doi.org/10.1109/DEVLRN.2013.6652554","url":null,"abstract":"In this work, we present a model based on social referencing mechanisms for the artwork appreciation development. We show that trying to reach autonomous robot artwork preferences is an interesting framework for developmental robotics. Therefore, we present an application of our model in a natural environment. The museum as an experiment place, helps us to benefit from a large number of visitors with different backgrounds and personal interests, as well as provides the challenge of the learning and recognition of an important number of artefacts (heterogeneous set of objects in terms of size, culture). To overcome some limitations like the autonomy in the learning, we proposed a first model for cumulative learning using a second-order conditioning approach with very encouraging results.","PeriodicalId":106997,"journal":{"name":"2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129416301","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：