Universal Approximation by Using the Correntropy Objective Function
نویسنده:
, , , , , , ,سال
: 2017
چکیده: Several objective functions have been proposed in the
literature to adjust the input parameters of a node in constructive
networks. Furthermore, many researchers have focused on the universal
approximation capability of the network based on the existing objective
functions. In this brief, we use a correntropy measure based on the
sigmoid kernel in the objective function to adjust the input parameters of
a newly added node in a cascade network. The proposed network is shown
to be capable of approximating any continuous nonlinear mapping with
probability one in a compact input sample space. Thus, the convergence is
guaranteed. The performance of our method was compared with that of
eight different objective functions, as well as with an existing one hidden
layer feedforward network on several real regression data sets with and
without impulsive noise. The experimental results indicate the benefits
of using a correntropy measure in reducing the root mean square error
and increasing the robustness to noise.
literature to adjust the input parameters of a node in constructive
networks. Furthermore, many researchers have focused on the universal
approximation capability of the network based on the existing objective
functions. In this brief, we use a correntropy measure based on the
sigmoid kernel in the objective function to adjust the input parameters of
a newly added node in a cascade network. The proposed network is shown
to be capable of approximating any continuous nonlinear mapping with
probability one in a compact input sample space. Thus, the convergence is
guaranteed. The performance of our method was compared with that of
eight different objective functions, as well as with an existing one hidden
layer feedforward network on several real regression data sets with and
without impulsive noise. The experimental results indicate the benefits
of using a correntropy measure in reducing the root mean square error
and increasing the robustness to noise.
کلیدواژه(گان): Cascade-correlation network (CNN),
correntropy,incremental constructive network,universal
approximation
کالکشن
:
-
آمار بازدید
Universal Approximation by Using the Correntropy Objective Function
Show full item record
| contributor author | مجتبی نیّری | en |
| contributor author | هادی صدوقی یزدی | en |
| contributor author | آلاله مسکوکی | en |
| contributor author | سیدمجتبی روحانی | en |
| contributor author | Mojtaba Nayyeri | fa |
| contributor author | Hadi Sadoghi Yazdi | fa |
| contributor author | Alaleh Maskooki | fa |
| contributor author | Modjtaba Rouhani | fa |
| date accessioned | 2020-06-06T13:37:38Z | |
| date available | 2020-06-06T13:37:38Z | |
| date issued | 2017 | |
| identifier uri | http://libsearch.um.ac.ir:80/fum/handle/fum/3362550 | |
| description abstract | Several objective functions have been proposed in the literature to adjust the input parameters of a node in constructive networks. Furthermore, many researchers have focused on the universal approximation capability of the network based on the existing objective functions. In this brief, we use a correntropy measure based on the sigmoid kernel in the objective function to adjust the input parameters of a newly added node in a cascade network. The proposed network is shown to be capable of approximating any continuous nonlinear mapping with probability one in a compact input sample space. Thus, the convergence is guaranteed. The performance of our method was compared with that of eight different objective functions, as well as with an existing one hidden layer feedforward network on several real regression data sets with and without impulsive noise. The experimental results indicate the benefits of using a correntropy measure in reducing the root mean square error and increasing the robustness to noise. | en |
| language | English | |
| title | Universal Approximation by Using the Correntropy Objective Function | en |
| type | Journal Paper | |
| contenttype | External Fulltext | |
| subject keywords | Cascade-correlation network (CNN) | en |
| subject keywords | correntropy | en |
| subject keywords | incremental constructive network | en |
| subject keywords | universal approximation | en |
| journal title | IEEE Transactions on Neural Networks and Learning Systems | fa |
| pages | 7-Jan | |
| journal volume | 2017 | |
| journal issue | 1 | |
| identifier link | https://profdoc.um.ac.ir/paper-abstract-1065662.html | |
| identifier articleid | 1065662 |