Zobrazeno 1 - 10
of 93
pro vyhledávání: '"Karhunen, Juha"'
Inspired by recent research, we explore ways to model the highly morphological Finnish language at the level of characters while maintaining the performance of word-level models. We propose a new Character-to-Word-to-Character (C2W2C) compositional l
Externí odkaz:
http://arxiv.org/abs/1612.03266
In this paper we investigate a link between state- space models and Gaussian Processes (GP) for time series modeling and forecasting. In particular, several widely used state- space models are transformed into continuous time form and corresponding G
Externí odkaz:
http://arxiv.org/abs/1610.08074
Deep convolutional neural networks (CNNs) have been immensely successful in many high-level computer vision tasks given large labeled datasets. However, for video semantic object segmentation, a domain where labels are scarce, effectively exploiting
Externí odkaz:
http://arxiv.org/abs/1606.02280
Autor:
Berglund, Mathias, Raiko, Tapani, Honkala, Mikko, Kärkkäinen, Leo, Vetek, Akos, Karhunen, Juha
Bidirectional recurrent neural networks (RNN) are trained to predict both in the positive and negative time directions simultaneously. They have not been used commonly in unsupervised tasks, because a probabilistic interpretation of the model has bee
Externí odkaz:
http://arxiv.org/abs/1504.01575
A software library for constructing and learning probabilistic models is presented. The library offers a set of building blocks from which a large variety of static and dynamic models can be built. These include hierarchical models for variances of o
Externí odkaz:
http://arxiv.org/abs/1207.1380
Publikováno v:
In Computers & Security September 2017 70:689-701
Publikováno v:
In Neurocomputing 3 August 2013 113:153-167
Publikováno v:
In Digital Signal Processing 2007 17(5):914-934
Autor:
Karhunen, Juha, Ukkonen, Tomas
Publikováno v:
In Neurocomputing 2007 70(16):2969-2979