Abstract
This dissertation consists of three main parts. In the first part, the existing methods of machine learning are applied to the environmental and astronomical datasets. The problems addressed in this part are the prediction of phosphorus concentration in the Pyhäjärvi lake (Finland) and the analysis of the correlation of geomagnetic storms with solar activity. For the first problem, several different models are built and the final accuracy is improved by variable selection and making an optimal ensemble. The second problem is solved by considering a correlation coefficient and estimating its uncertainty by the bootstrap method.
The second part of the dissertation is devoted to studying randomlyweighted neural networks or in more narrow terminology Extreme Learning Machines (ELM). Vanilla ELM is trained by ordinary linear regression. As a consequence, ELM has reasonable accuracy but its training is much faster than the training of other neural networks. In this dissertation ELM for time series forecasting is investigated. It is shown that Optimally Pruned ELM (OPELM) algorithm in combination with a certain prediction strategy is better than a baseline model for time series data from different domains. Besides, the general regression algorithm (Inc)OPELM is proposed which is significantly faster than the original OPELM but has the same performance.
Finally, in the third part of the dissertation, the probabilistic models for time series data are studied. Two types of probabilistic time series models are considered: linear statespace models and temporal Gaussian processes (GP). The connections between them are studied and new Gaussian process covariance functions are derived. These new covariance functions correspond to statespace models which are popular in the literature. Temporal Gaussian processes can be converted to statespace form as well. It is shown that this conversion allows expressing the inference in temporal GPs as operations with blocktridiagonal matrices. These matrix operations can be computed in linear time with respect to the number of samples or in sublinear time if parallel algorithms are utilized. Algorithms developed in this dissertation can serve as a basis for more complex models like spatiotemporal models and models with nonGaussian likelihoods.
Original language  English 

Qualification  Doctor's degree 
Awarding Institution 

Supervisors/Advisors 

Publisher  
Print ISBNs  9789526086705 
Electronic ISBNs  9789526086712 
Publication status  Published  2019 
MoE publication type  G5 Doctoral dissertation (article) 
Keywords
 randomlyweighted neural networs
 extreme learning machines
 Gaussian processes
 time series prediction
 statespace models
Fingerprint Dive into the research topics of 'Advances in RandomlyWeighted Neural Networks and Temporal Gaussian Processes'. Together they form a unique fingerprint.
Cite this
Grigorievskiy, A. (2019). Advances in RandomlyWeighted Neural Networks and Temporal Gaussian Processes. Aalto University.