Spatio-temporal event association using reward-modulated spike-time-dependent plasticity
For goal-directed learning in spiking neural networks, target spike templates are usually required.Optimal performance is achieved by minimising the error between the desired and output spike timings.However, in some dynamic environments, a set of learning targets with precise encoding is not always...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
Elsevier B.V.
2018
|
| Subjects: |
| _version_ | 1825805093299027968 |
|---|---|
| author | Yusoff, Nooraini Ibrahim, Mohammed Fadhil |
| author_facet | Yusoff, Nooraini Ibrahim, Mohammed Fadhil |
| author_sort | Yusoff, Nooraini |
| collection | UUM |
| description | For goal-directed learning in spiking neural networks, target spike templates are usually required.Optimal performance is achieved by minimising the error between the desired and output spike timings.However, in some dynamic environments, a set of learning targets with precise encoding is not always available.For this study, we associate a pair of spatio-temporal events with a target response using a reinforcement learning approach.The learning is implemented in a recurrent spiking neural network using reward-modulated spike-time-dependent plasticity.The learning protocol is simple and inspired by a behavioural experiment from a neuropsychology study.For a goal-directed application, learning does not require a target spike template.In this study, convergence is measured by synchronicity of activities in associated neuronal groups.As a result of learning, a network is able to associate a pair of events with a temporal delay in a dynamic setting. The results demonstrate that the algorithm can also learn temporal sequence detection.Learning has also been tested in face-voice association using real biometric data.The loose dependency between the model's anatomical properties and functionalities could offer a wide range of applications, especially in complex learning environments. |
| first_indexed | 2024-07-04T06:26:14Z |
| format | Article |
| id | uum-24363 |
| institution | Universiti Utara Malaysia |
| last_indexed | 2024-07-04T06:26:14Z |
| publishDate | 2018 |
| publisher | Elsevier B.V. |
| record_format | eprints |
| spelling | uum-243632018-07-04T01:51:42Z https://repo.uum.edu.my/id/eprint/24363/ Spatio-temporal event association using reward-modulated spike-time-dependent plasticity Yusoff, Nooraini Ibrahim, Mohammed Fadhil LB Theory and practice of education For goal-directed learning in spiking neural networks, target spike templates are usually required.Optimal performance is achieved by minimising the error between the desired and output spike timings.However, in some dynamic environments, a set of learning targets with precise encoding is not always available.For this study, we associate a pair of spatio-temporal events with a target response using a reinforcement learning approach.The learning is implemented in a recurrent spiking neural network using reward-modulated spike-time-dependent plasticity.The learning protocol is simple and inspired by a behavioural experiment from a neuropsychology study.For a goal-directed application, learning does not require a target spike template.In this study, convergence is measured by synchronicity of activities in associated neuronal groups.As a result of learning, a network is able to associate a pair of events with a temporal delay in a dynamic setting. The results demonstrate that the algorithm can also learn temporal sequence detection.Learning has also been tested in face-voice association using real biometric data.The loose dependency between the model's anatomical properties and functionalities could offer a wide range of applications, especially in complex learning environments. Elsevier B.V. 2018 Article PeerReviewed Yusoff, Nooraini and Ibrahim, Mohammed Fadhil (2018) Spatio-temporal event association using reward-modulated spike-time-dependent plasticity. Information Sciences, 451-45. pp. 143-160. ISSN 00200255 http://doi.org/10.1016/j.ins.2018.03.043 doi:10.1016/j.ins.2018.03.043 doi:10.1016/j.ins.2018.03.043 |
| spellingShingle | LB Theory and practice of education Yusoff, Nooraini Ibrahim, Mohammed Fadhil Spatio-temporal event association using reward-modulated spike-time-dependent plasticity |
| title | Spatio-temporal event association using reward-modulated spike-time-dependent plasticity |
| title_full | Spatio-temporal event association using reward-modulated spike-time-dependent plasticity |
| title_fullStr | Spatio-temporal event association using reward-modulated spike-time-dependent plasticity |
| title_full_unstemmed | Spatio-temporal event association using reward-modulated spike-time-dependent plasticity |
| title_short | Spatio-temporal event association using reward-modulated spike-time-dependent plasticity |
| title_sort | spatio temporal event association using reward modulated spike time dependent plasticity |
| topic | LB Theory and practice of education |
| work_keys_str_mv | AT yusoffnooraini spatiotemporaleventassociationusingrewardmodulatedspiketimedependentplasticity AT ibrahimmohammedfadhil spatiotemporaleventassociationusingrewardmodulatedspiketimedependentplasticity |