Mikko Lehtokangas

49 papers B 14C 1Journal 22Unranked 11

Year	Rank	Type	Title / Venue / Authors
2004	J	`jnl`	Complex RPROP-algorithm for neural network equalization of GSM data bursts. Neurocomputing Arto Kantsila, Mikko Lehtokangas, Jukka Saarinen
2004	J	`jnl`	Estimating movements of a robotic manipulator by hybrid constructive neural networks. Neurocomputing Jani J. T. Lahnajärvi, Mikko Lehtokangas, Jukka Saarinen
2003	—	`conf`	On center locating in RBF network equalization in the GSM system. NNSP Arto Kantsila, Mikko Lehtokangas, Jukka Saarinen
2003	B	`conf`	On radial basis function network equalization in the GSM system. ESANN Arto Kantsila, Mikko Lehtokangas, Jukka Saarinen
2002	J	`jnl`	Evaluation of constructive neural networks with cascaded architectures. Neurocomputing Jani J. T. Lahnajärvi, Mikko Lehtokangas, Jukka Saarinen
2002	J	`jnl`	Fuzzy detection of EEG alpha without amplitude thresholding. Artif. Intell. Medicine Eero Huupponen, Sari-Leena Himanen, Alpo Värri, Joel Hasan, Antti Saastamoinen, Mikko Lehtokangas, Jukka Saarinen
2000	—	`conf`	Adaptive channel equalization using classification trees. EUSIPCO Taneli Haverinen, Arto Kantsila, Mikko Lehtokangas, Jukka Saarinen
2000	—	`conf`	Comparison of fuzzy reasoning and Autoassociative MLP in sleep spindle detection. EUSIPCO Eero Huupponen, Alpo Värri, Joel Hasan, Sari-Leena Himanen, Mikko Lehtokangas, Jukka Saarinen
2000	J	`jnl`	Determining the number of centroids for CMLP network. Neural Networks Mikko Lehtokangas
2000	J	`jnl`	Fast Learning Using Multilayer Perceptron Networks with Adaptive Centroid Layer. Int. J. Pattern Recognit. Artif. Intell. Mikko Lehtokangas
2000	J	`jnl`	Modified cascade-correlation learning for classification. IEEE Trans. Neural Networks Learn. Syst. Mikko Lehtokangas
2000	J	`jnl`	Modified constructive backpropagation for regression. Neurocomputing Mikko Lehtokangas
2000	C	`conf`	On equalization with maximum covariance initialized cascade-correlation learning. ISCAS Arto Kantsila, Mikko Lehtokangas, Jukka Saarinen
2000	—	`conf`	Pattern recognition with novel support vector machine learning method. EUSIPCO Mikko Lehtokangas
2000	—	`conf`	Scatter compensation in digital radiography. EUSIPCO Vesa Onnia, Vesa Varjonen, Mari Lehtimaki, Mikko Lehtokangas, Jukka Saarinen
2000	—	`conf`	Support vector method for minutiae detection in fingerprint images. EUSIPCO Adrian Burian, Marius Tico, Mikko Lehtokangas, Pauli Kuosmanen, Jukka Saarinen
1999	B	`conf`	Adaptive training methods for optimal margin classification. IJCNN Mikko Lehtokangas
1999	J	`jnl`	Constructive Backpropagation for Recurrent Networks. Neural Process. Lett. Mikko Lehtokangas
1999	J	`jnl`	Fast initialization for cascade-correlation learning. IEEE Trans. Neural Networks Mikko Lehtokangas
1999	J	`jnl`	Feedforward Neural Network with Adaptive Reference Pattern Layer. Int. J. Neural Syst. Mikko Lehtokangas
1999	B	`conf`	Improved CBP learning with output bias decomposition. IJCNN Mikko Lehtokangas
1999	B	`conf`	Inverse PCA method for weight initialization in CMLP network. IJCNN Mikko Lehtokangas
1999	—	`conf`	MLP network for enhancement of noisy MFCC vectors. EUROSPEECH Hemmo Haverinen, Petri Salmela, Juha Häkkinen, Mikko Lehtokangas, Jukka Saarinen
1999	J	`jnl`	Modelling with constructive backpropagation. Neural Networks Mikko Lehtokangas
1999	J	`jnl`	Neural Network Based Digit Recognition System for Voice Dialling in Noisy Environments. Inf. Sci. Petri Salmela, Mikko Lehtokangas, Jukka Saarinen
1999	B	`conf`	On optimal usage of centroid units in CMLP network. IJCNN Mikko Lehtokangas
1999	B	`conf`	On weight initialization in cascade-correlation learning. IJCNN Mikko Lehtokangas
1999	—	`conf`	Robust speaker adaptation of continuous density HMMS using multilayer perceptron network. EUROSPEECH Mikko Harju, Petri Salmela, Olli Viikki, Mikko Lehtokangas, Jukka Saarinen
1999	B	`conf`	Time series modelling with recurrent CBP. IJCNN Mikko Lehtokangas
1998	J	`jnl`	Centroid based Multilayer Perceptron Networks. Neural Process. Lett. Mikko Lehtokangas, Jukka Saarinen
1998	B	`conf`	Fast Training with Maximum Covariance Initialized Centroid MLP. ICONIP Mikko Lehtokangas
1998	—	`conf`	Improved neural network equalization by the use of maximum covariance weight initialization. EUSIPCO Arto Kantsila, Mikko Lehtokangas, Jukka Saarinen
1998	B	`conf`	Initialization of MLP with Adaptive Neuron Specific Preprocessing Units. ICONIP Mikko Lehtokangas
1998	J	`jnl`	Modeling of speech signals using an optimal neural network structure based on the PMDL principle. IEEE Trans. Speech Audio Process. Xiao Ming Gao, Seppo J. Ovaska, Mikko Lehtokangas, Jukka Saarinen
1998	B	`conf`	Noisy Digit Recognition with MLP and HMMS Using Modified Cost Function in MCE Training. ICONIP Petri Salmela, Mikko Lehtokangas, Jukka Saarinen
1998	B	`conf`	On Training of MLP with Adaptive Neuron Specific Preprocessing Units. ICONIP Mikko Lehtokangas
1998	—	`conf`	Reference Pattern Weight Initialization for Equalization. IEA/AIE (Vol. 2) Mikko Lehtokangas
1998	J	`jnl`	Waveform detection with RBF network - Application to automated EEG analysis. Neurocomputing Antti Saastamoinen, Timo Pietilä, Alpo Värri, Mikko Lehtokangas, Jukka Saarinen
1998	—	`ch.`	Weight initialization techniques. Algorithms and Architectures Mikko Lehtokangas, Jukka Saarinen, Petri Salmela, Kimmo Kaski
1998	J	`jnl`	Weight initialization with reference patterns. Neurocomputing Mikko Lehtokangas, Jukka Saarinen
1997	—	`conf`	Discriminative Isolated Digit Recognition in Noisy Environments Using Multilayer Perceptron Networks. ICONIP (2) Petri Salmela, Kari Laurila, Mikko Lehtokangas, Jukka Saarinen
1997	J	`jnl`	Short Term Electric Load Forecasting Using A Neural Network with Fuzzy Hidden Neurons. Neural Comput. Appl. Seppo Kuusisto, Mikko Lehtokangas, Jukka Saarinen, Kimmo Kaski
1996	B	`conf`	Maximum covariance method for weight initialization of multilayer perceptron network. ESANN Mikko Lehtokangas, Petri Korpisaari, Kimmo Kaski
1996	J	`jnl`	Predictive minimum description length criterion for time series modeling with neural networks. Neural Comput. Mikko Lehtokangas, Jukka Saarinen, Pentti Huuhtanen, Kimmo Kaski
1995	J	`jnl`	Accelerating training of radial basis function networks with Cascade-Correlation algorithm. Neurocomputing Mikko Lehtokangas, Jukka Saarinen, Kimmo Kaski
1995	J	`jnl`	Fine-tuning Cascade-Correlation trained feedforward network with backpropagation. Neural Process. Lett. Mikko Lehtokangas, Jukka Saarinen, Kimmo Kaski
1995	J	`jnl`	Initializing Weights of a Multilayer Perceptron Network by Using the Orthogonal Least Squares Algorithm. Neural Comput. Mikko Lehtokangas, Jukka Saarinen, Kimmo Kaski, Pentti Huuhtanen
1995	B	`conf`	Simplified cascade-correlation learning. ESANN Mikko Lehtokangas, Jukka Saarinen, Kimmo Kaski
1993	B	`conf`	Neural Network Optimization Tool Based on Predictive MDL Principle for Time Series Prediction. ICTAI Mikko Lehtokangas, Jukka Saarinen, Pentti Huuhtanen, Kimmo Kaski