Showing 1–13 of 13 results for author: Gianniotis, N

A Gaussian process cross-correlation approach to time delay estimation in active galactic nuclei

Authors: F. Pozo Nuñez, N. Gianniotis, K. L. Polsterer

Abstract: We present a probabilistic cross-correlation approach to estimate time delays in the context of reverberation map** (RM) of Active Galactic Nuclei (AGN). We reformulate the traditional interpolated cross-correlation method as a statistically principled model that delivers a posterior distribution for the delay. The method employs Gaussian processes as a model for observed AGN light curves. We de… ▽ More We present a probabilistic cross-correlation approach to estimate time delays in the context of reverberation map** (RM) of Active Galactic Nuclei (AGN). We reformulate the traditional interpolated cross-correlation method as a statistically principled model that delivers a posterior distribution for the delay. The method employs Gaussian processes as a model for observed AGN light curves. We describe the mathematical formalism and demonstrate the new approach using both simulated light curves and available RM observations. The proposed method delivers a posterior distribution for the delay that accounts for observational noise and the non-uniform sampling of the light curves. This feature allow us to fully quantify its uncertainty and propagate it to subsequent calculations of dependent physical quantities, e.g., black hole masses. It delivers out-of-sample predictions, which enables us to subject it to model selection and it can calculate the joint posterior delay for more than two light curves. Because of the numerous advantages of our reformulation and the simplicity of its application, we anticipate that our method will find favour not only in the specialised community of RM, but in all fields where cross-correlation analysis is performed. We provide the algorithms and examples of their application as part of our Julia GPCC package. △ Less

Submitted 11 April, 2023; originally announced April 2023.

Comments: 13 pages, 16 figures, Accepted for publication in Astronomy and Astrophysics

Journal ref: A&A 674, A83 (2023)

Disentangling the optical AGN and Host-galaxy luminosity with a probabilistic Flux Variation Gradient

Authors: N. Gianniotis, F. Pozo Nuñez, K. L. Polsterer

Abstract: We present a novel Probabilistic Flux Variation Gradient (PFVG) approach to to separate the contributions of active galactic nuclei (AGN) and host galaxies in the context of photometric reverberation map** (PRM) of AGN. We explored the ability of recovering the fractional contribution in a model-independent way using the entire set of light curves obtained through different filters and photometr… ▽ More We present a novel Probabilistic Flux Variation Gradient (PFVG) approach to to separate the contributions of active galactic nuclei (AGN) and host galaxies in the context of photometric reverberation map** (PRM) of AGN. We explored the ability of recovering the fractional contribution in a model-independent way using the entire set of light curves obtained through different filters and photometric apertures simultaneously. The method is based on the observed bluer when brighter phenomenon that is attributed to the superimposition of a two-component structure; the red host galaxy, which is constant in time, and the varying blue AGN. We describe the PFVG mathematical formalism and demonstrate its performance using simulated light curves and available PRM observations. The new probabilistic approach is able to recover host-galaxy fluxes to within 1% precision as long as the light curves do not show a significant contribution from time delays. This represents a significant improvement with respect to previous applications of the traditional FVG method to PRM data. The proposed PFVG provides an efficient and accurate way to separate the AGN and host-galaxy luminosities in PRM monitoring data. The method will be especially helpful in the case of large upcoming photometric survey telescopes such as the public optical/near-infrared Legacy Survey of Space and Time (LSST) at the Vera C. Rubin Observatory. Finally, we have made the algorithms freely available as part of our Julia PFVG package. △ Less

Submitted 28 October, 2021; v1 submitted 7 September, 2021; originally announced September 2021.

Comments: Accepted for publication in Astronomy and Astrophysics

Journal ref: A&A 657, A126 (2022)

Optical continuum photometric reverberation map** of the Seyfert-1 galaxy Mrk509

Authors: F. Pozo Nuñez, N. Gianniotis, J. Blex, T. Lisow, R. Chini, K. L. Polsterer, J. -U. Pott, J. Esser, G. Pietrzyński

Abstract: We present the results of a two year optical continuum photometric reverberation map** campaign carried out on the nucleus of the Seyfert-1 galaxy Mrk509. Specially designed narrow-band filters were used in order to mitigate the line and pseudo-continuum contamination of the signal from the broad line region, while allowing for high-accuracy flux-calibration over a large field of view. We obtain… ▽ More We present the results of a two year optical continuum photometric reverberation map** campaign carried out on the nucleus of the Seyfert-1 galaxy Mrk509. Specially designed narrow-band filters were used in order to mitigate the line and pseudo-continuum contamination of the signal from the broad line region, while allowing for high-accuracy flux-calibration over a large field of view. We obtained light curves with a sub-day time sampling and typical flux uncertainties of $1\%$. The high photometric precision allowed us to measure inter-band continuum time delays of up to $\sim 2$ days across the optical range. The time delays are consistent with the relation $τ\propto λ^{4/3}$ predicted for an optically thick and geometrically thin accretion disk model. The size of the disk is, however, a factor of 1.8 larger than predictions based on the standard thin-disk theory. We argue that, for the particular case of Mrk509, a larger black hole mass due to the unknown geometry scaling factor can reconcile the difference between the observations and theory. △ Less

Submitted 21 December, 2019; originally announced December 2019.

Comments: 20 pages, 11 figures, published on Monthly Notices of the Royal Astronomical Society

Journal ref: Monthly Notices of the Royal Astronomical Society, Volume 490, Issue 3, December 2019, Pages 3936-3951

Approximate Variational Inference Based on a Finite Sample of Gaussian Latent Variables

Authors: Nikolaos Gianniotis, Christoph Schnörr, Christian Molkenthin, Sanjay Singh Bora

Abstract: Variational methods are employed in situations where exact Bayesian inference becomes intractable due to the difficulty in performing certain integrals. Typically, variational methods postulate a tractable posterior and formulate a lower bound on the desired integral to be approximated, e.g. marginal likelihood. The lower bound is then optimised with respect to its free parameters, the so called v… ▽ More Variational methods are employed in situations where exact Bayesian inference becomes intractable due to the difficulty in performing certain integrals. Typically, variational methods postulate a tractable posterior and formulate a lower bound on the desired integral to be approximated, e.g. marginal likelihood. The lower bound is then optimised with respect to its free parameters, the so called variational parameters. However, this is not always possible as for certain integrals it is very challenging (or tedious) to come up with a suitable lower bound. Here we propose a simple scheme that overcomes some of the awkward cases where the usual variational treatment becomes difficult. The scheme relies on a rewriting of the lower bound on the model log-likelihood. We demonstrate the proposed scheme on a number of synthetic and real examples, as well as on a real geophysical model for which the standard variational approaches are inapplicable. △ Less

Submitted 11 June, 2019; originally announced June 2019.

Comments: published in Pattern Analysis and Applications

Journal ref: Pattern Anal Applic (2016) 19: 475

Efficient Optimization of Echo State Networks for Time Series Datasets

Authors: Jacob Reinier Maat, Nikos Gianniotis, Pavlos Protopapas

Abstract: Echo State Networks (ESNs) are recurrent neural networks that only train their output layer, thereby precluding the need to backpropagate gradients through time, which leads to significant computational gains. Nevertheless, a common issue in ESNs is determining its hyperparameters, which are crucial in instantiating a well performing reservoir, but are often set manually or using heuristics. In th… ▽ More Echo State Networks (ESNs) are recurrent neural networks that only train their output layer, thereby precluding the need to backpropagate gradients through time, which leads to significant computational gains. Nevertheless, a common issue in ESNs is determining its hyperparameters, which are crucial in instantiating a well performing reservoir, but are often set manually or using heuristics. In this work we optimize the ESN hyperparameters using Bayesian optimization which, given a limited budget of function evaluations, outperforms a grid search strategy. In the context of large volumes of time series data, such as light curves in the field of astronomy, we can further reduce the optimization cost of ESNs. In particular, we wish to avoid tuning hyperparameters per individual time series as this is costly; instead, we want to find ESNs with hyperparameters that perform well not just on individual time series but rather on groups of similar time series without sacrificing predictive performance significantly. This naturally leads to a notion of clusters, where each cluster is represented by an ESN tuned to model a group of time series of similar temporal behavior. We demonstrate this approach both on synthetic datasets and real world light curves from the MACHO survey. We show that our approach results in a significant reduction in the number of ESN models required to model a whole dataset, while retaining predictive performance for the series in each cluster. △ Less

Submitted 12 March, 2019; originally announced March 2019.

Journal ref: 2018 International Joint Conference on Neural Networks (IJCNN), pp. 1-7. IEEE, 2018

Mixed Variational Inference

Authors: Nikolaos Gianniotis

Abstract: The Laplace approximation has been one of the workhorses of Bayesian inference. It often delivers good approximations in practice despite the fact that it does not strictly take into account where the volume of posterior density lies. Variational approaches avoid this issue by explicitly minimising the Kullback-Leibler divergence DKL between a postulated posterior and the true (unnormalised) logar… ▽ More The Laplace approximation has been one of the workhorses of Bayesian inference. It often delivers good approximations in practice despite the fact that it does not strictly take into account where the volume of posterior density lies. Variational approaches avoid this issue by explicitly minimising the Kullback-Leibler divergence DKL between a postulated posterior and the true (unnormalised) logarithmic posterior. However, they rely on a closed form DKL in order to update the variational parameters. To address this, stochastic versions of variational inference have been devised that approximate the intractable DKL with a Monte Carlo average. This approximation allows calculating gradients with respect to the variational parameters. However, variational methods often postulate a factorised Gaussian approximating posterior. In doing so, they sacrifice a-posteriori correlations. In this work, we propose a method that combines the Laplace approximation with the variational approach. The advantages are that we maintain: applicability on non-conjugate models, posterior correlations and a reduced number of free variational parameters. Numerical experiments demonstrate improvement over the Laplace approximation and variational inference with factorised Gaussian posteriors. △ Less

Submitted 28 February, 2022; v1 submitted 15 January, 2019; originally announced January 2019.

Comments: Conference submission, 15 pages

Journal ref: IJCNN 2019

Modelling multimodal photometric redshift regression with noisy observations

Authors: S. D. Kügler, N. Gianniotis

Abstract: In this work, we are trying to extent the existing photometric redshift regression models from modeling pure photometric data back to the spectra themselves. To that end, we developed a PCA that is capable of describing the input uncertainty (including missing values) in a dimensionality reduction framework. With this "spectrum generator" at hand, we are capable of treating the redshift regression… ▽ More In this work, we are trying to extent the existing photometric redshift regression models from modeling pure photometric data back to the spectra themselves. To that end, we developed a PCA that is capable of describing the input uncertainty (including missing values) in a dimensionality reduction framework. With this "spectrum generator" at hand, we are capable of treating the redshift regression problem in a fully Bayesian framework, returning a posterior distribution over the redshift. This approach allows therefore to approach the multimodal regression problem in an adequate fashion. In addition, input uncertainty on the magnitudes can be included quite naturally and lastly, the proposed algorithm allows in principle to make predictions outside the training values which makes it a fascinating opportunity for the detection of high-redshifted quasars. △ Less

Submitted 20 July, 2016; originally announced July 2016.

Comments: 7 pages. arXiv admin note: text overlap with arXiv:1606.06094

A Spectral Model for Multimodal Redshift Estimation

Authors: Sven D. Kugler, Nikolaos Gianniotis, Kai L. Polsterer

Abstract: We present a physically inspired model for the problem of redshift estimation. Typically, redshift estimation has been treated as a regression problem that takes as input magnitudes and maps them to a single target redshift. In this work we acknowledge the fact that observed magnitudes may actually admit multiple plausible redshifts, i.e. the distribution of redshifts explaining the observed magni… ▽ More We present a physically inspired model for the problem of redshift estimation. Typically, redshift estimation has been treated as a regression problem that takes as input magnitudes and maps them to a single target redshift. In this work we acknowledge the fact that observed magnitudes may actually admit multiple plausible redshifts, i.e. the distribution of redshifts explaining the observed magnitudes (or colours) is multimodal. Hence, employing one of the standard regression models, as is typically done, is insufficient for this kind of problem, as most models implement either one-to-one or many-to-one map**s. The observed multimodality of solutions is a direct consequence of (a) the variety of physical mechanisms that give rise to the observations, (b) the limited number of measurements available and (c) the presence of noise in photometric measurements. Our proposed solution consists in formulating a model from first principles capable of generating spectra. The generated spectra are integrated over filter curves to produce magnitudes which are then matched to the observed magnitudes. The resulting model naturally expresses a multimodal posterior over possible redshifts, includes measurement uncertainty (e.g. missing values) and is shown to perform favourably on a real dataset. △ Less

Submitted 20 June, 2016; originally announced June 2016.

Model-Coupled Autoencoder for Time Series Visualisation

Authors: Nikolaos Gianniotis, Sven D. Kügler, Peter Tiňo, Kai L. Polsterer

Abstract: We present an approach for the visualisation of a set of time series that combines an echo state network with an autoencoder. For each time series in the dataset we train an echo state network, using a common and fixed reservoir of hidden neurons, and use the optimised readout weights as the new representation. Dimensionality reduction is then performed via an autoencoder on the readout weight rep… ▽ More We present an approach for the visualisation of a set of time series that combines an echo state network with an autoencoder. For each time series in the dataset we train an echo state network, using a common and fixed reservoir of hidden neurons, and use the optimised readout weights as the new representation. Dimensionality reduction is then performed via an autoencoder on the readout weight representations. The crux of the work is to equip the autoencoder with a loss function that correctly interprets the reconstructed readout weights by associating them with a reconstruction error measured in the data space of sequences. This essentially amounts to measuring the predictive performance that the reconstructed readout weights exhibit on their corresponding sequences when plugged back into the echo state network with the same fixed reservoir. We demonstrate that the proposed visualisation framework can deal both with real valued sequences as well as binary sequences. We derive magnification factors in order to analyse distance preservations and distortions in the visualisation space. The versatility and advantages of the proposed method are demonstrated on datasets of time series that originate from diverse domains. △ Less

Submitted 21 January, 2016; originally announced January 2016.

An Explorative Approach for Inspecting Kepler Data

Authors: S. D. Kügler, N. Gianniotis, K. L. Polsterer

Abstract: The Kepler survey has provided a wealth of astrophysical knowledge by continuously monitoring over 150,000 stars. The resulting database contains thousands of examples of known variability types and at least as many that cannot be classified yet. In order to reveal the knowledge hidden in the database, we introduce a new visualisation method that allows us to inspect time series exploratively. To… ▽ More The Kepler survey has provided a wealth of astrophysical knowledge by continuously monitoring over 150,000 stars. The resulting database contains thousands of examples of known variability types and at least as many that cannot be classified yet. In order to reveal the knowledge hidden in the database, we introduce a new visualisation method that allows us to inspect time series exploratively. To that end, we propose dimensionality reduction on the parameters of a model capable of representing time series as fixed-length vector representation. We show that a more refined objective function can be chosen by minimising the prediction error of the data reconstruction instead of the reconstruction of the model parameters. The proposed visualisation exhibits a strong correlation between the variability behaviour of the light curves and their physical properties. As a consequence, temperature and surface gravity can, for some stars, be directly inferred from non- (or quasi-) periodic light curves. △ Less

Submitted 4 November, 2015; v1 submitted 14 August, 2015; originally announced August 2015.

Comments: 7 pages, 8 figures, accepted for publication in MNRAS

Autoencoding Time Series for Visualisation

Authors: Nikolaos Gianniotis, Dennis Kügler, Peter Tino, Kai Polsterer, Ranjeev Misra

Abstract: We present an algorithm for the visualisation of time series. To that end we employ echo state networks to convert time series into a suitable vector representation which is capable of capturing the latent dynamics of the time series. Subsequently, the obtained vector representations are put through an autoencoder and the visualisation is constructed using the activations of the bottleneck. The cr… ▽ More We present an algorithm for the visualisation of time series. To that end we employ echo state networks to convert time series into a suitable vector representation which is capable of capturing the latent dynamics of the time series. Subsequently, the obtained vector representations are put through an autoencoder and the visualisation is constructed using the activations of the bottleneck. The crux of the work lies with defining an objective function that quantifies the reconstruction error of these representations in a principled manner. We demonstrate the method on synthetic and real data. △ Less

Submitted 5 May, 2015; originally announced May 2015.

Comments: Published in ESANN 2015

Featureless Classification of Light Curves

Authors: Sven Dennis Kügler, Nikos Gianniotis, Kai Lars Polsterer

Abstract: In the era of rapidly increasing amounts of time series data, classification of variable objects has become the main objective of time-domain astronomy. Classification of irregularly sampled time series is particularly difficult because the data cannot be represented naturally as a vector which can be directly fed into a classifier. In the literature, various statistical features serve as vector r… ▽ More In the era of rapidly increasing amounts of time series data, classification of variable objects has become the main objective of time-domain astronomy. Classification of irregularly sampled time series is particularly difficult because the data cannot be represented naturally as a vector which can be directly fed into a classifier. In the literature, various statistical features serve as vector representations. In this work, we represent time series by a density model. The density model captures all the information available, including measurement errors. Hence, we view this model as a generalisation to the static features which directly can be derived, e.g., as moments from the density. Similarity between each pair of time series is quantified by the distance between their respective models. Classification is performed on the obtained distance matrix. In the numerical experiments, we use data from the OGLE and ASAS surveys and demonstrate that the proposed representation performs up to par with the best cur- rently used feature-based approaches. The density representation preserves all static information present in the observational data, in contrast to a less complete description by features. The density representation is an upper boundary in terms of information made available to the classifier. Consequently, the predictive power of the proposed classification depends on the choice of similarity measure and classifier, only. Due to its principled nature, we advocate that this new approach of representing time series has potential in tasks beyond classification, e.g., unsupervised learning. △ Less

Submitted 20 May, 2015; v1 submitted 17 April, 2015; originally announced April 2015.

Comments: Accepted for publication in MNRAS

Topographic Map** of astronomical light curves via a physically inspired Probabilistic model

Authors: Nikolaos Gianniotis, Peter Tino, Steve Spreckley, Somak Raychaudhury

Abstract: We present a probabilistic generative approach for constructing topographic maps of light curves from eclipsing binary stars. The model defines a low-dimensional manifold of local noise models induced by a smooth non-linear map** from a low-dimensional latent space into the space of probabilistic models of the observed light curves. The local noise models are physical models that describe how… ▽ More We present a probabilistic generative approach for constructing topographic maps of light curves from eclipsing binary stars. The model defines a low-dimensional manifold of local noise models induced by a smooth non-linear map** from a low-dimensional latent space into the space of probabilistic models of the observed light curves. The local noise models are physical models that describe how such light curves are generated. Due to the principled probabilistic nature of the model, a cost function arises naturally and the model parameters are fitted via MAP estimation using the Expectation-Maximisation algorithm. Once the model has been trained, each light curve may be projected to the latent space as the the mean posterior probability over the local noise models. We demonstrate our approach on a dataset of artificially generated light curves and on a dataset comprised of light curves from real observations. △ Less

Submitted 21 September, 2009; originally announced September 2009.

Comments: 10 pages, 5 figures. Accepted for the (refereed) proceedings of the International Conference on Artificial Neural Networks (ICANN 2009), September 14-17, Limassol, Cyprus (http://www.kios.org.cy/ICANN09/)