arXiv subjects

Fredrik Sandin

Publications and source records attributed to Fredrik Sandin.

29 records · Page 2Linked to original sources

arXiv2020

Improving Image Autoencoder Embeddings with Perceptual Loss

Autoencoders are commonly trained using element-wise loss. However, element-wise loss disregards high-level structures in the image which can lead to embeddings that disregard them as well. A recent improvement to autoencoders that helps alleviate this problem is the use of perceptual loss. This work investigates perceptual loss from the perspective of encoder embeddings themselves. Autoencoders are trained to embed images from three different computer vision datasets using perceptual loss based on a pretrained model as well as pixel-wise loss. A host of different predictors are trained to perform object positioning and classification on the datasets given the embedded images as input. The two kinds of losses are evaluated by comparing how the predictors performed with embeddings from the differently trained autoencoders. The results show that, in the image domain, the embeddings generated by autoencoders trained with perceptual loss enable more accurate predictions than those trained with element-wise loss. Furthermore, the results show that, on the task of object positioning of a small-scale feature, perceptual loss can improve the results by a factor 10. The experimental setup is available online: https://github.com/guspih/Perceptual-Autoencoders

cs.CV↗

arXiv2019

Dictionary learning approach to monitoring of wind turbine drivetrain bearings

Condition monitoring is central to the efficient operation of wind farms due to the challenging operating conditions, rapid technology development and large number of aging wind turbines. In particular, predictive maintenance planning requires the early detection of faults with few false positives. Achieving this type of detection is a challenging problem due to the complex and weak signatures of some faults, particularly the faults that occur in some of the drivetrain bearings. Here, we investigate recently proposed condition monitoring methods based on unsupervised dictionary learning using vibration data recorded over 46 months under typical industrial operations. Thus, we contribute novel test results and real world data that are made publicly available. The results of former studies addressing condition monitoring tasks using dictionary learning indicate that unsupervised feature learning is useful for diagnosis and anomaly detection purposes. However, these studies are based on small sets of labeled data from test rigs operating under controlled conditions that focus on classification tasks, which are useful for quantitative method comparisons but gives little insight into how useful these approaches are in practice. In this study, dictionaries are learned from gearbox vibrations in six different turbines, and the dictionaries are subsequently propagated over a few years of monitoring data when faults are known to occur. We perform the experiment using two different sparse coding algorithms to investigate if the algorithm selected affects the features of abnormal conditions. We calculate the dictionary distance between the initial and propagated dictionaries and find the time periods of abnormal dictionary adaptation starting six months before a drivetrain bearing replacement and one year before the resulting gearbox replacement.

eess.SP↗

arXiv2019

Synaptic Delays for Temporal Feature Detection in Dynamic Neuromorphic Processors

Spiking neural networks implemented in dynamic neuromorphic processors are well suited for spatiotemporal feature detection and learning, for example in ultra low-power embedded intelligence and deep edge applications. Such pattern recognition networks naturally involve a combination of dynamic delay mechanisms and coincidence detection. Inspired by an auditory feature detection circuit in crickets, featuring a delayed excitation by postinhibitory rebound, we investigate disynaptic delay elements formed by inhibitory-excitatory pairs of dynamic synapses. We configure such disynaptic delay elements in the DYNAP-SE neuromorphic processor and characterize the distribution of delayed excitations resulting from device mismatch. Furthermore, we present a network that mimics the auditory feature detection circuit of crickets and demonstrate how varying synapse weights, input noise and processor temperature affects the circuit. Interestingly, we find that the disynaptic delay elements can be configured such that the timing and magnitude of the delayed postsynaptic excitation depend mainly on the efficacy of the inhibitory and excitatory synapses, respectively. Delay elements of this kind can be implemented in other reconfigurable dynamic neuromorphic processors and opens up for synapse level temporal feature tuning with large fan-in and flexible delays of order 10-100 ms.

cs.NE↗

arXiv2019

Interoperability and machine-to-machine translation model with mappings to machine learning tasks

Modern large-scale automation systems integrate thousands to hundreds of thousands of physical sensors and actuators. Demands for more flexible reconfiguration of production systems and optimization across different information models, standards and legacy systems challenge current system interoperability concepts. Automatic semantic translation across information models and standards is an increasingly important problem that needs to be addressed to fulfill these demands in a cost-efficient manner under constraints of human capacity and resources in relation to timing requirements and system complexity. Here we define a translator-based operational interoperability model for interacting cyber-physical systems in mathematical terms, which includes system identification and ontology-based translation as special cases. We present alternative mathematical definitions of the translator learning task and mappings to similar machine learning tasks and solutions based on recent developments in machine learning. Possibilities to learn translators between artefacts without a common physical context, for example in simulations of digital twins and across layers of the automation pyramid are briefly discussed.

cs.LG↗

arXiv2016

Dictionary Learning with Equiprobable Matching Pursuit

Sparse signal representations based on linear combinations of learned atoms have been used to obtain state-of-the-art results in several practical signal processing applications. Approximation methods are needed to process high-dimensional signals in this way because the problem to calculate optimal atoms for sparse coding is NP-hard. Here we study greedy algorithms for unsupervised learning of dictionaries of shift-invariant atoms and propose a new method where each atom is selected with the same probability on average, which corresponds to the homeostatic regulation of a recurrent convolutional neural network. Equiprobable selection can be used with several greedy algorithms for dictionary learning to ensure that all atoms adapt during training and that no particular atom is more likely to take part in the linear combination on average. We demonstrate via simulation experiments that dictionary learning with equiprobable selection results in higher entropy of the sparse representation and lower reconstruction and denoising errors, both in the case of ordinary matching pursuit and orthogonal matching pursuit with shift-invariant dictionaries. Furthermore, we show that the computational costs of the matching pursuits are lower with equiprobable selection, leading to faster and more accurate dictionary learning algorithms.

cs.LG↗

arXiv2015

Towards zero-configuration condition monitoring based on dictionary learning

Condition-based predictive maintenance can significantly improve overall equipment effectiveness provided that appropriate monitoring methods are used. Online condition monitoring systems are customized to each type of machine and need to be reconfigured when conditions change, which is costly and requires expert knowledge. Basic feature extraction methods limited to signal distribution functions and spectra are commonly used, making it difficult to automatically analyze and compare machine conditions. In this paper, we investigate the possibility to automate the condition monitoring process by continuously learning a dictionary of optimized shift-invariant feature vectors using a well-known sparse approximation method. We study how the feature vectors learned from a vibration signal evolve over time when a fault develops within a ball bearing of a rotating machine. We quantify the adaptation rate of learned features and find that this quantity changes significantly in the transitions between normal and faulty states of operation of the ball bearing.

cs.CV↗

arXiv2011

Incremental dimension reduction of tensors with random index

We present an incremental, scalable and efficient dimension reduction technique for tensors that is based on sparse random linear coding. Data is stored in a compactified representation with fixed size, which makes memory requirements low and predictable. Component encoding and decoding are performed on-line without computationally expensive re-analysis of the data set. The range of tensor indices can be extended dynamically without modifying the component representation. This idea originates from a mathematical model of semantic memory and a method known as random indexing in natural language processing. We generalize the random-indexing algorithm to tensors and present signal-to-noise-ratio simulations for representations of vectors and matrices. We present also a mathematical analysis of the approximate orthogonality of high-dimensional ternary vectors, which is a property that underpins this and other similar random-coding approaches to dimension reduction. To further demonstrate the properties of random indexing we present results of a synonym identification task. The method presented here has some similarities with random projection and Tucker decomposition, but it performs well at high dimensionality only (n>10^3). Random indexing is useful for a range of complex practical problems, e.g., in natural language processing, data mining, pattern recognition, event detection, graph searching and search engines. Prototype software is provided. It supports encoding and decoding of tensors of order >= 1 in a unified framework, i.e., vectors, matrices and higher order tensors.

cs.DS↗

arXiv2010

Have neutron stars a dark matter core?

Recent observational results for the masses and radii of some neutron stars are in contrast with typical observations and theoretical predictions for "normal" neutron stars. We propose that their unusual properties can be interpreted as the signature of a dark matter core inside them. This interpretation requires that the dark matter is made of some form of stable, long-living or in general non-annihilating particles, that can accumulate in the star. In the proposed scenario all mass-radius measurements can be explained with one nuclear matter equation of state and a dark core of varying relative size. This hypothesis will be challenged by forthcoming observations and could eventually be a useful tool for the determination of dark matter.

astro-ph.HE↗

arXiv2009

Effects of mirror dark matter on neutron stars

If dark matter is made of mirror baryons, they are present in all gravitationally bound structures. Here we investigate some effects of mirror dark matter on neutron stars and discuss possible observational consequences. The general-relativistic hydrostatic equations are generalized to spherical objects with multiple fluids that interact by gravity. We use the minimal parity-symmetric extension of the standard model, which implies that the microphysics is the same in the two sectors. We find that the mass-radius relation is significantly modified in the presence of a few percent mirror baryons. This effect mimics that of other exotica, e.g., quark matter. In contrast to the common view that the neutron-star equilibrium sequence is unique, we show that it depends on the relative number of mirror baryons to ordinary baryons. It is therefore history dependent. The critical mass for core collapse, i.e., the process by which neutron stars are created, is modified in the presence of mirror baryons. We calculate the modified Chandrasekhar mass and fit it with a polynomial. A few percent mirror baryons is sufficient to lower the critical mass for core collapse by ~0.1 M_sun. This could allow for the formation of extraordinary compact neutron stars with low mass.

astro-ph↗

arXiv2008

Single-flavor CSL phase in compact stars

We suggest a scenario where the three light quark flavors are sequentially deconfined under increasing pressure in cold asymmetric nuclear matter as e.g. in neutron stars. The basis for our analysis is a chiral quark matter model of Nambu--Jona-Lasinio (NJL) type with diquark pairing in the spin-1 single flavor (CSL), spin-0 two flavor (2SC) and three flavor (CFL) channels. We find that nucleon dissociation sets in at about the saturation density, n_0, when the down-quark Fermi sea is populated (d-quark dripline) due to the flavor asymmetry induced by beta-equilibrium and charge neutrality. At about 3n_0 u-quarks appear and a two-flavor color superconducting (2SC) phase is formed. The s-quark Fermi sea is populated only at still higher baryon density, when the quark chemical potential is of the order of the dynamically generated strange quark mass. We construct two different hybrid equations of state (EoS) using the Dirac-Brueckner Hartree-Fock (DBHF) approach and the EoS by Shen et al in the nuclear matter sector. The corresponding hybrid star sequences have maximum masses of, respectively, 2.1 and 2.0 M_solar. Two- and three-flavor quark-matter phases exist only in gravitationally unstable hybrid star solutions in the DBHF case, while the Shen-based EoS produce stable configurations with a 2SC phase component in the core of massive stars. Nucleon dissociation due to d-quark drip at the crust-core boundary fulfills basic criteria for a deep crustal heating process which is required to explain superbusts as well as cooling of X-ray transients.

astro-ph↗

arXiv2008

1-2-3-flavor color superconductivity in compact stars

We suggest a scenario where the three light quark flavors are sequentially deconfined under increasing pressure in cold asymmetric nuclear matter, e.g., as in neutron stars. The basis for our analysis is a chiral quark matter model of Nambu--Jona-Lasinio (NJL) type with diquark pairing in the spin-1 single flavor (CSL) and spin-0 two/three flavor (2SC/CFL) channels, and a Dirac-Brueckner Hartree-Fock (DBHF) approach in the nuclear matter sector. We find that nucleon dissociation sets in at about the saturation density, n_0, when the down-quark Fermi sea is populated (d-quark dripline) due to the flavor asymmetry imposed by beta-equilibrium and charge neutrality. At about 3n_0 u-quarks appear forming a two-flavor color superconducting (2SC) phase, while the s-quark Fermi sea is populated only at still higher baryon density. The hybrid star sequence has a maximum mass of 2.1 M_sun. Two- and three-flavor quark matter phases are found only in gravitationally unstable hybrid star solutions.

nucl-th↗