Mikko Lehtokangas - researchr alias

researchr

You are not signed in
Sign in
Sign up

Viewing Publication 1 - 49 from 49

2004

Complex RPROP-algorithm for neural network equalization of GSM data burstsArto Kantsila, Mikko Lehtokangas, Jukka Saarinen. ijon, 61:339-360, 2004. [doi]

Estimating movements of a robotic manipulator by hybrid constructive neural networksJani J. T. Lahnajärvi, Mikko Lehtokangas, Jukka Saarinen. ijon, 56:345-363, 2004. [doi]

2003

On radial basis function network equalization in the GSM systemArto Kantsila, Mikko Lehtokangas, Jukka Saarinen. esann 2003: 179-184 [doi]

On center locating in RBF network equalization in the GSM systemArto Kantsila, Mikko Lehtokangas, Jukka Saarinen. nnsp 2003: 575-584 [doi]

2002

Fuzzy detection of EEG alpha without amplitude thresholdingEero Huupponen, Sari-Leena Himanen, Alpo Värri, Joel Hasan, Antti Saastamoinen, Mikko Lehtokangas, Jukka Saarinen. artmed, 24(2):133-147, 2002.

Evaluation of constructive neural networks with cascaded architecturesJani J. T. Lahnajärvi, Mikko Lehtokangas, Jukka Saarinen. ijon, 48(1-4):573-607, 2002. [doi]

2000

Support vector method for minutiae detection in fingerprint imagesAdrian Burian, Marius Tico, Mikko Lehtokangas, Pauli Kuosmanen, Jukka Saarinen. eusipco 2000: 1-4 [doi]

Adaptive channel equalization using classification treesTaneli Haverinen, Arto Kantsila, Mikko Lehtokangas, Jukka Saarinen. eusipco 2000: 1-4 [doi]

Comparison of fuzzy reasoning and Autoassociative MLP in sleep spindle detectionEero Huupponen, Alpo Värri, Joel Hasan, Sari-Leena Himanen, Mikko Lehtokangas, Jukka Saarinen. eusipco 2000: 1-4 [doi]

On equalization with maximum covariance initialized cascade-correlation learningArto Kantsila, Mikko Lehtokangas, Jukka Saarinen. iscas 2000: 168-171 [doi]

Determining the number of centroids for CMLP networkMikko Lehtokangas. NN, 13(4-5):525-531, 2000. [doi]

Fast Learning Using Multilayer Perceptron Networks with Adaptive Centroid LayerMikko Lehtokangas. ijprai, 14(2):211-224, 2000.

Modified constructive backpropagation for regressionMikko Lehtokangas. ijon, 35(1-4):113-122, 2000. [doi]

Modified cascade-correlation learning for classificationMikko Lehtokangas. tnn, 11(3):795-798, 2000. [doi]

Pattern recognition with novel support vector machine learning methodMikko Lehtokangas. eusipco 2000: 1-4 [doi]

Scatter compensation in digital radiographyVesa Onnia, Vesa Varjonen, Mari Lehtimaki, Mikko Lehtokangas, Jukka Saarinen. eusipco 2000: 1-4 [doi]

1999

Robust speaker adaptation of continuous density HMMS using multilayer perceptron networkMikko Harju, Petri Salmela, Olli Viikki, Mikko Lehtokangas, Jukka Saarinen. interspeech 1999: 2499-2502 [doi]

MLP network for enhancement of noisy MFCC vectorsHemmo Haverinen, Petri Salmela, Juha Häkkinen, Mikko Lehtokangas, Jukka Saarinen. interspeech 1999: 2371-2374 [doi]

Fast initialization for cascade-correlation learningMikko Lehtokangas. tnn, 10(2):410-414, 1999. [doi]

Modelling with constructive backpropagationMikko Lehtokangas. NN, 12(4-5):707-716, 1999. [doi]

Feedforward Neural Network with Adaptive Reference Pattern LayerMikko Lehtokangas. IJNS, 9(1):1-9, 1999. [doi]

Constructive Backpropagation for Recurrent NetworksMikko Lehtokangas. npl, 9(3):271-278, 1999.

On weight initialization in cascade-correlation learningMikko Lehtokangas. ijcnn 1999: 1471-1474 [doi]

Improved CBP learning with output bias decompositionMikko Lehtokangas. ijcnn 1999: 1706-1709 [doi]

Adaptive training methods for optimal margin classificationMikko Lehtokangas. ijcnn 1999: 1415-1418 [doi]

Time series modelling with recurrent CBPMikko Lehtokangas. ijcnn 1999: 2560-2563 [doi]

On optimal usage of centroid units in CMLP networkMikko Lehtokangas. ijcnn 1999: 1604-1607 [doi]

Inverse PCA method for weight initialization in CMLP networkMikko Lehtokangas. ijcnn 1999: 878-881 [doi]

Neural Network Based Digit Recognition System for Voice Dialling in Noisy EnvironmentsPetri Salmela, Mikko Lehtokangas, Jukka Saarinen. isci, 121(3-4):171-199, 1999.

1998

Modeling of speech signals using an optimal neural network structure based on the PMDL principleXiao Ming Gao, Seppo J. Ovaska, Mikko Lehtokangas, Jukka Saarinen. taslp, 6(2):177-180, 1998. [doi]

Improved neural network equalization by the use of maximum covariance weight initializationArto Kantsila, Mikko Lehtokangas, Jukka Saarinen. eusipco 1998: 1-4 [doi]

Centroid based Multilayer Perceptron NetworksMikko Lehtokangas, Jukka Saarinen. npl, 7(2):101-106, 1998.

On Training of MLP with Adaptive Neuron Specific Preprocessing UnitsMikko Lehtokangas. iconip 1998: 1567-1570

Weight initialization with reference patternsMikko Lehtokangas, Jukka Saarinen. ijon, 20(1-3):265-278, 1998. [doi]

Reference Pattern Weight Initialization for EqualizationMikko Lehtokangas. ieaaie 1998: 574-581

Fast Training with Maximum Covariance Initialized Centroid MLPMikko Lehtokangas. iconip 1998: 1385-1388

Initialization of MLP with Adaptive Neuron Specific Preprocessing UnitsMikko Lehtokangas. iconip 1998: 1571-1574

Weight initialization techniquesMikko Lehtokangas, Jukka Saarinen, Petri Salmela, Kimmo Kaski. In Cornelius T. Leondes, editor, Algorithms and Architectures. Volume 1 of Neural network systems techniques and applications, pages 87-121, Morgan Kaufmann, 1998. [doi]

Waveform detection with RBF network - Application to automated EEG analysisAntti Saastamoinen, Timo Pietilä, Alpo Värri, Mikko Lehtokangas, Jukka Saarinen. ijon, 20(1-3):1-13, 1998. [doi]

Noisy Digit Recognition with MLP and HMMS Using Modified Cost Function in MCE TrainingPetri Salmela, Mikko Lehtokangas, Jukka Saarinen. iconip 1998: 1118-1121

1997

Short Term Electric Load Forecasting Using A Neural Network with Fuzzy Hidden NeuronsSeppo Kuusisto, Mikko Lehtokangas, Jukka Saarinen, Kimmo Kaski. nca, 6(1):42-56, 1997.

Discriminative Isolated Digit Recognition in Noisy Environments Using Multilayer Perceptron NetworksPetri Salmela, K. Laurila, Mikko Lehtokangas, Jukka Saarinen. iconip 1998: 1116-1119

1996

Predictive minimum description length criterion for time series modeling with neural networksMikko Lehtokangas, Jukka Saarinen, Pentti Huuhtanen, Kimmo Kaski. neco, 8(3):583-593, 1996.

Maximum covariance method for weight initialization of multilayer perceptron networkMikko Lehtokangas, Petri Korpisaari, Kimmo Kaski. esann 1996: [doi]

1995

1993

Neural Network Optimization Tool Based on Predictive MDL Principle for Time Series PredictionMikko Lehtokangas, Jukka Saarinen, Pentti Huuhtanen, Kimmo Kaski. ictai 1993: 338-342

All Publications

Filter by Year
OR AND NOT 1

Filter by Type
OR AND NOT 1

Filter by Tag
OR AND NOT 1

Filter by Venue

[+]
OR AND NOT 1

Filter by Author

[+]
OR AND NOT 1

Filter by Top Terms

[+]
OR AND NOT 1

Researchr

Publications by 'Mikko Lehtokangas'

Viewing Publication 1 - 49 from 49

2004

2003

2002

2000

1999

1998

1997

1996

1995

1993

All Publications

Filter by YearOR AND NOT 1

Filter by TypeOR AND NOT 1

Filter by TagOR AND NOT 1

Filter by Venue [+]OR AND NOT 1

Filter by Author [+]OR AND NOT 1

Filter by Top Terms [+]OR AND NOT 1

Researchr

Publications by 'Mikko Lehtokangas'

Viewing Publication 1 - 49 from 49

2004

2003

2002

2000

1999

1998

1997

1996

1995

1993

Filter by Year
OR AND NOT 1

Filter by Type
OR AND NOT 1

Filter by Tag
OR AND NOT 1

Filter by Venue

[+]
OR AND NOT 1

Filter by Author

[+]
OR AND NOT 1

Filter by Top Terms

[+]
OR AND NOT 1