OFDM
Relay
Cognitive Radio
Multiple Antennas
Resource Allocation
Full Duplex
Spectrum Sensing
Synchronization
Spectrum Sharing
Interference Cancellation
Channel Estimation
Feedback
Heterogeneous Networks
Bi-directional
Energy Harvesting
Stochastic Geometry
HetNet
relay networks
FBMC
Equalization
channel capacity
TVWS
CDMA
interference
in-band full-duplex system
Duplex
MIMO
Ultra Low Power
C-V2V
5G
Reliability
SC-FDMA
interference suppression
D-TDD
CLI
indoor positioning
reinforcement learning
RSRP weighting
- Computation offloading
smart factory
Cell-free
multi-access edge computing
estimated position overlapping
—Device-to-device (D2D)
estimated position updating
mMIMO
control overhead
hybrid
NR positioning
Femtocell
Rat-dependent positioning
frame structure
Zigbee
body area networks
channel estimation error
Handoff
CoMP
User grouping
power uncertainty
ultra-dense small cell network
mode selection
antenna arrays
5G mobile communication
UFMC
resource block management
inter user interference
WVAN
health care
partial overlap
GFDM
Dynamic TDD
Multi-user Receiver
Number of training blocks
Uplink SCMA system
V2X
Vehicular communication
cross-link interference
LTE-TDD
FS-NOMA
Location-based
user fairness
Mode 3
QR Factorization
Metaheuristics
P-NOMA
non-orthogonal multiple access
dynamic HetNet
spectrum partitioning
and 5G networks.
massive connectivity
non-orthogonal multiple access (NOMA)
overloading
DQN
OTDOA
distributed mode
Communication range
resource selection
maximum likelihood method
Resource sharing
Power allocation
packet delay
Sooyong Choi, KyunByoung Ko and Daesik Hong, "A Multilayer Feedforward Neural Network Having N/4 nodes in Two Hidden Layers", IJCNN, Jun. 2001
[IEEE IJCNN]
조회 64866
Status : | Presented |
---|---|
Date : | 2001-07 |
Title : | A Multilayer Feedforward Neural Network Having N/4 nodes in Two Hidden Layers |
Authors : | Sooyong Choi, KyunByoung Ko and Daesik Hong |
Conference : | IEEE International Joint Conference on Neural Networks |
Abstract : | In order to reduce the complexity of a single hidden layer multilayer neural network, a new two hidden layer MFNN (THL-MFNN) with a combined structure of a RBFN and MLPs is proposed, and its associated training method is discussed. The proposed THL-MFNN can be easily constructed, and can be efficiently trained by online recursive methods. The performance of the proposed THL-MFNN with P/4+2=18 hidden nodes and 34 weights is equal to that of an optimum Bayesian equalizer using an RBFN with P=64 hidden nodes and 64 weights. The role of each layer in the proposed THL-MFNN is presented by a theoretical approach, and the feasibility of a more reduced structure is given |
URL : | http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=938413 |
Download : | http://mirinae.yonsei.ac.kr/?module=file...94d708d23e |
Sooyong Choi; Kyunbyoung Ko; Daesik Hong; , "A multilayer feedforward neural
network having N/4 nodes in two hidden layers," Neural Networks, 2001.
Proceedings. IJCNN '01. International Joint Conference on , vol.3, no.,
pp.1675-1680 vol.3, 2001
doi: 10.1109/IJCNN.2001.938413
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=938413&isnumber=20319
카테고리
- 전체(260)
- [ICNN] (3)
- [ICEIC] (15)
- [ITC-CSCC] (12)
- [JTC-CSCC] (3)
- [IEEE ICASSP] (10)
- [IEEE ICC] (22)
- [IEEE ICCE] (1)
- [IEEE ICCS] (3)
- [IEEE ICOIN] (3)
- [IEEE ICONIP] (2)
- [IEEE IJCNN] (9)
- [IEEE INTERMAG] (8)
- [IEEE ISIT] (1)
- [IEEE GLOBECOM] (21)
- [IEEE MILCOM] (11)
- [IEEE PIMRC] (17)
- [IEEE VTC] (73)
- [IEEE WCNC] (8)
- [IEEE TENCON] (1)
- [IEEE CNCC] (1)
- [Other Conf. Papers] (36)