his.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Durán, Boris
Alternative names
Publications (8 of 8) Show all publications
Duran, B. & Sandamirskaya, Y. (2018). Learning Temporal Intervals in Neural Dynamics. IEEE Transactions on Cognitive and Developmental Systems, 10(2), 359-372
Open this publication in new window or tab >>Learning Temporal Intervals in Neural Dynamics
2018 (English)In: IEEE Transactions on Cognitive and Developmental Systems, ISSN 2379-8920, Vol. 10, no 2, p. 359-372Article in journal (Refereed) Published
National Category
Human Computer Interaction
Research subject
Interaction Lab (ILAB)
Identifiers
urn:nbn:se:his:diva-13431 (URN)10.1109/TCDS.2017.2676839 (DOI)000435198600020 ()2-s2.0-85048698309 (Scopus ID)
Available from: 2017-03-13 Created: 2017-03-13 Last updated: 2018-07-05Bibliographically approved
Thill, S., Duran, B. & Hemeren, P. (Eds.). (2013). Social signals in action recognition and intention understanding. Frontiers
Open this publication in new window or tab >>Social signals in action recognition and intention understanding
2013 (English)Collection (editor) (Refereed)
Place, publisher, year, edition, pages
Frontiers, 2013
National Category
Psychology Social Psychology Computer and Information Sciences
Identifiers
urn:nbn:se:his:diva-9354 (URN)
Available from: 2014-06-02 Created: 2014-06-02 Last updated: 2019-05-21Bibliographically approved
Durán, B., Sandamirskaya, Y. & Schöner, G. (2012). A dynamic field architecture for the generation of hierarchically organized sequences. In: Alessandro E. P. Villa, Włodzisław Duch, Péter Érdi, Francesco Masulli, Günther Palm (Ed.), Artificial Neural Networks and Machine Learning – ICANN 2012: 22nd International Conference on Artificial Neural Networks, Lausanne, Switzerland, September 11-14, 2012, Proceedings, Part I. Paper presented at 22nd International Conference on Artificial Neural Networks, ICANN 2012, 11 September 2012 through 14 September 2012, Lausanne (pp. 25-32). Springer Berlin/Heidelberg (PART 1)
Open this publication in new window or tab >>A dynamic field architecture for the generation of hierarchically organized sequences
2012 (English)In: Artificial Neural Networks and Machine Learning – ICANN 2012: 22nd International Conference on Artificial Neural Networks, Lausanne, Switzerland, September 11-14, 2012, Proceedings, Part I / [ed] Alessandro E. P. Villa, Włodzisław Duch, Péter Érdi, Francesco Masulli, Günther Palm, Springer Berlin/Heidelberg, 2012, no PART 1, p. 25-32Conference paper, Published paper (Refereed)
Abstract [en]

A dilemma arises when sequence generation is implemented on embodied autonomous agents. While achieving an individual action goal, the agent must be in a stable state to link to fluctuating and time-varying sensory inputs. To transition to the next goal, the previous state must be released from stability. A previous proposal of a neural dynamics solved this dilemma by inducing an instability when a "condition of satisfaction" signals that an action goal has been reached. The required structure of dynamic coupling limited the complexity and flexibility of sequence generation, however. We address this limitation by showing how the neural dynamics can be generalized to generate hierarchically structured behaviors. Directed couplings downward in the hierarchy initiate chunks of actions, directed couplings upward in the hierarchy signal their termination. We analyze the mathematical mechanisms and demonstrate the flexibility of the scheme in simulation. © 2012 Springer-Verlag.

Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2012
Series
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), ISSN 0302-9743 ; 7552 LNCS
Keywords
Dynamic Field Theory, Hierarchies, Intentionality, Sequences, Dynamic couplings, Dynamic fields, Mathematical mechanisms, Neural dynamics, Sensory input, Sequence generation, Stable state, Time varying, Autonomous agents, Couplings, Dynamics, Neural networks
National Category
Computer and Information Sciences
Research subject
Technology
Identifiers
urn:nbn:se:his:diva-8370 (URN)10.1007/978-3-642-33269-2_4 (DOI)2-s2.0-84867686740 (Scopus ID)978-3-642-33268-5 (ISBN)978-3-642-33269-2 (ISBN)
Conference
22nd International Conference on Artificial Neural Networks, ICANN 2012, 11 September 2012 through 14 September 2012, Lausanne
Available from: 2013-08-08 Created: 2013-08-08 Last updated: 2018-01-11Bibliographically approved
Durán, B., Lee, G. & Lowe, R. (2012). Learning a DFT-based sequence with reinforcement learning: A NAO implementation. Paladyn - Journal of Behavioral Robotics, 3(4), 181-187
Open this publication in new window or tab >>Learning a DFT-based sequence with reinforcement learning: A NAO implementation
2012 (English)In: Paladyn - Journal of Behavioral Robotics, ISSN 2080-9778, E-ISSN 2081-4836, Vol. 3, no 4, p. 181-187Article in journal (Refereed) Published
Place, publisher, year, edition, pages
Springer, 2012
Keywords
sequences, neural dynamics, reinforcement learning, humanoid
National Category
Computer Vision and Robotics (Autonomous Systems)
Identifiers
urn:nbn:se:his:diva-10147 (URN)10.2478/s13230-013-0109-5 (DOI)
Available from: 2014-10-29 Created: 2014-10-29 Last updated: 2018-01-11Bibliographically approved
Durán, B. & Thill, S. (2012). Modelling interaction in multi-modal affordance processing with neural dynamics. In: Tom Ziemke, Christian Balkenius, John Hallam (Ed.), From Animals to Animats 12: 12th International Conference on Simulation of Adaptive Behavior, SAB 2012, Odense, Denmark, August 27-30, 2012. Proceedings. Paper presented at 12th International Conference on Simulation of Adaptive Behavior, SAB 2012; Odense; 27 August 2012 through 30 August 2012 (pp. 75-84). Berlin, Heidelberg: Springer Berlin/Heidelberg
Open this publication in new window or tab >>Modelling interaction in multi-modal affordance processing with neural dynamics
2012 (English)In: From Animals to Animats 12: 12th International Conference on Simulation of Adaptive Behavior, SAB 2012, Odense, Denmark, August 27-30, 2012. Proceedings / [ed] Tom Ziemke, Christian Balkenius, John Hallam, Berlin, Heidelberg: Springer Berlin/Heidelberg, 2012, p. 75-84Conference paper, Published paper (Refereed)
Abstract [en]

Behavioral studies on the activation of affordances by understanding observation and action sentences on graspable objects show a direct relationship between the canonical orientation of graspable objects, their dimension and the kind of grip required by those objects to be grasped. The present work introduces the concepts of Dynamic Field Theory for modeling the results observed in the behavioral studies previously mentioned. The model was not only able to replicate qualitatively similar results regarding reaction times, but also the identification of same versus different object and the distinction between observable versus action sentences. The model shows the potential of dynamic field theory for the design and implementation of brain inspired cognitive systems. © 2012 Springer-Verlag.

Place, publisher, year, edition, pages
Berlin, Heidelberg: Springer Berlin/Heidelberg, 2012
Series
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), ISSN 0302-9743 ; 7426 LNAI
Keywords
affordances, Dynamic Field Theory, modelling, Neural dynamics
National Category
Computer and Information Sciences
Research subject
Technology
Identifiers
urn:nbn:se:his:diva-7040 (URN)10.1007/978-3-642-33093-3_8 (DOI)2-s2.0-84866023831 (Scopus ID)978-3-642-33092-6 (ISBN)978-3-642-33093-3 (ISBN)
Conference
12th International Conference on Simulation of Adaptive Behavior, SAB 2012; Odense; 27 August 2012 through 30 August 2012
Available from: 2013-01-23 Created: 2013-01-23 Last updated: 2018-01-11Bibliographically approved
Li, C., Lowe, R., Duran, B. & Ziemke, T. (2011). Humanoids that crawl: Comparing gait performance of iCub and NAO using a CPG architecture. In: Shaozi Li, Ying Dai (Ed.), Proceedings - 2011 IEEE International Conference on Computer Science and Automation Engineering, CSAE 2011: . Paper presented at 2011 IEEE International Conference on Computer Science and Automation Engineering, CSAE 2011, 10 June 2011 through 12 June 2011, Shanghai (pp. 577-582). IEEE conference proceedings
Open this publication in new window or tab >>Humanoids that crawl: Comparing gait performance of iCub and NAO using a CPG architecture
2011 (English)In: Proceedings - 2011 IEEE International Conference on Computer Science and Automation Engineering, CSAE 2011 / [ed] Shaozi Li, Ying Dai, IEEE conference proceedings, 2011, p. 577-582Conference paper, Published paper (Refereed)
Abstract [en]

In this article, a generic CPG architecture is used to model infant crawling gaits and is implemented on the NAO robot platform. The CPG architecture is chosen via a systematic approach to designing CPG networks on the basis of group theory and dynamic systems theory. The NAO robot performance is compared to the iCub robot which has a different anatomical structure. Finally, the comparison of performance and NAO whole-body stability are assessed to show the adaptive property of the CPG architecture and the extent of its ability to transfer to different robot morphologies. © 2011 IEEE.

Place, publisher, year, edition, pages
IEEE conference proceedings, 2011
Series
Proceedings - 2011 IEEE International Conference on Computer Science and Automation Engineering, CSAE 2011 ; 4
Keywords
CPG, Crawling, iCub, Infant development, NAO, Algebra, Network architecture, Robots, System theory, Computer architecture
Identifiers
urn:nbn:se:his:diva-7262 (URN)10.1109/CSAE.2011.5952916 (DOI)2-s2.0-80051897170 (Scopus ID)978-1-4244-8725-7 (ISBN)978-1-4244-8728-8 (ISBN)978-1-4244-8727-1 (ISBN)978-1-4244-8726-4 (ISBN)
Conference
2011 IEEE International Conference on Computer Science and Automation Engineering, CSAE 2011, 10 June 2011 through 12 June 2011, Shanghai
Available from: 2013-02-19 Created: 2013-02-19 Last updated: 2017-12-20Bibliographically approved
Thill, S., Hemeren, P. E. & Durán, B. (2011). Prediction of human action segmentation based on end-effector kinematics using linear models. In: Kokinov, B. et al. (Ed.), European Perspectives on Cognitive Science: Proceedings of the European Conference on Cognitive Science (pp. 6 sidor). Sofia: New Bulgarian University Press
Open this publication in new window or tab >>Prediction of human action segmentation based on end-effector kinematics using linear models
2011 (English)In: European Perspectives on Cognitive Science: Proceedings of the European Conference on Cognitive Science / [ed] Kokinov, B. et al., Sofia: New Bulgarian University Press , 2011, p. 6 sidor-Conference paper, Published paper (Refereed)
Abstract [en]

The work presented in this paper builds on previous research which analysed human action segmentation in the case of simple object manipulations with the hand (rather than larger-scale actions). When designing algorithms to segment observed actions, for instance to train robots by imitation, the typical approach involves non-linear models but it is less clear whether human action segmentation is also based on such analyses. In the present paper, we therefore explore (1) whether linear models built from observed kinematic variables of a human hand can accurately predict human action segmentation and (2) what kinematic variables are the most important in such a task. In previous work, we recorded speed, acceleration and change in direction for the wrist and the tip of each of the five fingers during the execution of actions as well as the segmentation of these actions into individual components by humans. Here, we use this data to train a large number of models based on every possible training set available and find that, amongst others, the speed of the wrist as well as the change in direction of the index finger were preferred in models with good performance. Overall, the best models achieved R2 values over 0.5 on novel test data but the average performance of trained models was modest. We suggest that this is due to a suboptimal training set (which was not specifically designed for the present task) and that further work be carried out to identify better training sets as our initial results indicate that linear models may indeed be a viable approach to predicting human action segmentation.

Place, publisher, year, edition, pages
Sofia: New Bulgarian University Press, 2011
Keywords
Action segmentation; Motion primitives; Linear model; Stepwise regression
National Category
Computer and Information Sciences
Research subject
Technology
Identifiers
urn:nbn:se:his:diva-5196 (URN)978-954-535-660-5 (ISBN)
Available from: 2011-07-04 Created: 2011-07-04 Last updated: 2018-01-12
Lowe, R., Duran, B. & Ziemke, T. (2010). A Dynamic Field Theoretic Model of Iowa Gambling Task Performance. In: 2010 IEEE 9th International Conference on Development and Learning (ICDL): Ann Arbor, MI, August 18-21, 2010. Paper presented at 2010 IEEE 9th International Conference on Development and Learning, Ann Arbor, MI, August 18-21, 2010 (pp. 297-304). IEEE conference proceedings
Open this publication in new window or tab >>A Dynamic Field Theoretic Model of Iowa Gambling Task Performance
2010 (English)In: 2010 IEEE 9th International Conference on Development and Learning (ICDL): Ann Arbor, MI, August 18-21, 2010, IEEE conference proceedings, 2010, p. 297-304Conference paper, Published paper (Refereed)
Abstract [en]

Choice behaviour where outcome-contingencies vary or are prohabilistic has been the focus of many benchmark tasks of infant to adult development in the psychology literature. Dynamic field theoretic (DFT) investigations of cognitive and behavioural competencies have been used in order to identify parameters critical to infant development. In this paper we report the findings of a DFT model that is able to replicate normal functioning adult  performance on the Iowa gambling task (IGT).  The model offers a simple demonstration proof of the parsimonious reversal learning alternative to Damasio’s somatic marker  explanation of IGT performance. Our simple model demonstrates a potentially important role for reinforcement/reward learning to generating behaviour that allows for advantageous performance. We compare our DFT modelling approach to one used on the A-not-B infant paradigm and suggest that a critical aspect of development lies in the ability to flexibly trade off perseverative versus exploratory behaviour in order to capture statistical choice-outcome contingencies. Finally, we discuss the importance of an investigation of the IGT in an embodied setting where reward prediction learning may provide critical means by which adaptive behavioural reversals can be enacted.

Place, publisher, year, edition, pages
IEEE conference proceedings, 2010
National Category
Computer and Information Sciences
Research subject
Technology
Identifiers
urn:nbn:se:his:diva-4809 (URN)10.1109/DEVLRN.2010.5578826 (DOI)2-s2.0-78149243222 (Scopus ID)978-1-4244-6902-4 (ISBN)1-4244-6902-3 (ISBN)978-1-4244-6900-0 (ISBN)1-4244-6900-7 (ISBN)
Conference
2010 IEEE 9th International Conference on Development and Learning, Ann Arbor, MI, August 18-21, 2010
Available from: 2011-04-12 Created: 2011-04-12 Last updated: 2018-01-12Bibliographically approved
Organisations

Search in DiVA

Show all publications