Aaltodoc
Aaltodoc is the institutional repository of Aalto University.
Aaltodoc has a new updated appearance. Instructions for searching and more information is available here.
Communities in Aaltodoc
Select a community to browse its collections.
Now showing 1 - 15 of 15
Recent Submissions
Helsingin kauppakorkeakoulu - vuosikertomus 1987
(Helsinki School of Economics, 1988) Leponiemi, Arvi
Helsingin Kauppakorkeakoulu - vuosikertomus 1991
(Helsinki School of Economics, 1992) Vaivio, Fedi
Sample-efficient inference for agent-based cognitive models and other computationally intensive simulators
(Aalto University, 2023) Aushev, Alexander; Tietotekniikan laitos; Department of Computer Science; Probabilistic Machine Learning; Perustieteiden korkeakoulu; School of Science; Kaski, Samuel, Aalto University, Finland and The University of Manchester, United Kingdom
In recent years, simulator models have become increasingly popular in many scientific domains, such as epidemiology, cosmology, and behavioural sciences. Since simulators often do not have tractable likelihoods, which are either too costly to evaluate or not available, the field needs to resort to likelihood-free inference (LFI), which uses forward simulations instead. With the development of more complex simulators, traditional LFI methods become unfeasible as the cost of simulations significantly increases. This thesis deals with three challenges that arise in the context of computationally heavy simulators and for which the existing LFI methods, such as approximate Bayesian computation, synthetic likelihood, or neural density estimation approaches, are inadequate since they require a large number of simulations.
The first challenge is modelling complex simulator noise, which influences the accuracy of LFI methods and becomes problematic when simulations are computationally costly. The existing methods either oversimplify the noise (e.g., by assuming it to be Gaussian) or require an infeasible number of simulations to accurately model it. We show how to handle multimodal, non-stationary, and heteroscedastic noise distributions in LFI while also assuming a small simulation budget. For this, we adopt deep Gaussian process surrogates in Bayesian Optimisation (BO), along with novel quantile-based multimodal-capable modifications for the acquisition function and posterior extraction procedures.
Another challenge for modern LFI approaches occurs when they are applied to time-series settings, as these methods either need an accurate model of transition dynamics available or always assume it to be linear. We propose a way of estimating the unknown transition dynamics for state predictions in simulator-based dynamical systems, which greatly reduces the required simulation budget and also enables time-series prediction. Our proposed approach uses a multi-objective surrogate for LFI and a semi-parametric model for the transition dynamics.
Finally, we significantly reduce the time required to select agent-based cognitive models with limited experimental designs. The previous methods have primarily focused on either model selection or parameter estimation, while we achieve both in a fraction of the time. This is accomplished through a novel simulator-based utility objective for selecting designs in BO and a LFI approximation of model marginal likelihood for model selection. This new method is needed for developing and verifying computational cognitive theories, which often lack tractable likelihoods.
Interpretable artificial neural networks for fMRI data classification
(Aalto University, 2023) Gotsopoulos, Athanasios; Sams, Mikko, Prof., Aalto University, Department of Neuroscience and Biomedical Engineering, Finland; Neurotieteen ja lääketieteellisen tekniikan laitos; Department of Neuroscience and Biomedical Engineering; Brain & Mind Laboratory; Perustieteiden korkeakoulu; School of Science; Lampinen, Jouko, Prof., Aalto University, Department of Computer Science, Finland
Functional magnetic resonance imaging (fMRI) technology allows non-invasive measurement of neuronal activity in the human brain with a combination of reasonable temporal and fine spatial resolution. Recently, multivariate methods have attracted attention in fMRI data analysis to study task-related activation patterns. Concurrent research in the field of machine learning has led to the establishment of inherently multivariate computational graphs that facilitate efficient, robust and interpretable classification of fMRI data.
Here we studied methods for classification of fMRI data based on neural networks. In particular, we focused on techniques that assess the contribution of different brain regions to the classification result, referred to as "importance maps" and proposed novel neuroscientifically motivated architectures.
In the first study, we successfully classified basic emotions from fMRI data, elicited by short movies and mental imagery, generating whole brain importance maps indicating the contribution of individual voxels to the classification result.
The second study provided a comparison of importance extraction methods and their reproducibility, applied to both simulated and real data sets, revealing patterns that do not convey significant univariate information.
The third study examined the effect of distractors in visual imagery using classification methods and importance map extraction, identifying robust activation patterns related to shape imagery and a visual distractor in object-selective lateral extrastriate cortex at the junction of left occipital, temporal and parietal lobes.
The fourth study examined the use of anatomically driven topologies based on spatial information. In particular, the addition of layers motivated by voxel proximity and brain atlases to the model, led to an increase in the classification accuracy and produced smoother and more interpretable importance maps.
The purpose of this thesis is to showcase machine learning techniques specifically designed for analyzing neuroscience data. This work aims to motivate further research towards the use of machine learning as a means to gain a better understanding of the human brain.
Photonic and Electronic Characterization of Two-dimensional Transition Metal Dichalcogenides
(Aalto University, 2023) Shafi, Abde Mayeen; Mackenzie, David, Dr., Kyocera Tikitin Oy, Finland; Elektroniikan ja nanotekniikan laitos; Department of Electronics and Nanoengineering; Nanoscience and Advanced Materials; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Lipsanen, Harri, Prof., Aalto University, Department of Electronics and Nanoengineering, Finland
Two-dimensional (2D) transition metal dichalcogenides (TMDCs) hold promise for numerous unprecedented applications in nanophotonics, optoelectronics, and nanoelectronics, owing to their extraordinary electrical and optical properties. However, these materials still face several challenges, including limited light-matter interactions, low luminescent yield, reduced carrier mobility, and susceptibility to environmental changes. This thesis aims to address the aforementioned limitations by employing various advanced techniques to enhance the optical and electronic properties of these materials.
In this thesis, the light-matter interaction in TMDCs is enhanced by realizing mixed-dimensional heterostructures. High-performance photonic and optoelectronic devices are constructed by investigating two distinct types of these heterostructures. Firstly, monolayer MoS2 is transferred onto AlGaAs nanowires to create a mixed-dimensional heterostructure. A significant enhancement in Raman and photoluminescence responses is achieved from the heterostructure attributed to the electromagnetic field confinement within the high refractive index nanowire. The heterostructure also exhibits optical anisotropy due to the 3-fold rotational symmetry breaking of MoS2 caused by the nanowire. Additionally, the fabricated phototransistor using this heterostructure demonstrates improved responsivity and detectivity. Secondly, another mixed-dimensional heterostructure is formed by epitaxially growing InP nanowires directly on MoS2. High-density nanowire growth is achieved while ensuring the stability of MoS2. This heterostructure generates strong second- and third-harmonic signals and, notably, 5th and 7th-order high-harmonic signals, opening up potential applications such as lasers and electro-optic modulators.
In the subsequent part of the thesis, the electronic properties of TMDCs are investigated and tuned to fabricate high-performance electronic and optoelectronic devices. At first, the impact of high temperatures on multilayer MoTe2 field-effect transistors is systematically explored to determine the optimal annealing temperature for the devices and acquire a deeper understanding of the surface oxidation-mediated defect formation and hopping transport mechanism in MoTe2 devices. Furthermore, a straightforward technique is proposed that involves substrate engineering and Al2O3 passivation to enhance the performance of few-layer MoTe2 devices by introducing local tensile strain and reducing electron-phonon scattering in the channel. This results in significant improvements in carrier mobility and device quality. Lastly, a simple optical writing technique is employed to transform the semiconducting 2H phase of MoTe2 into the metallic 1T´ phase, resulting in improved third harmonic generation signals and the performance of optoelectronic devices. These findings show great promise for advancing integrated photonic and optoelectronic circuits based on 2D-TMDCs.
