Showing 1–19 of 19 results for author: Zimmermann, H

VISA: Variational Inference with Sequential Sample-Average Approximations

Authors: Heiko Zimmermann, Christian A. Naesseth, Jan-Willem van de Meent

Abstract: We present variational inference with sequential sample-average approximation (VISA), a method for approximate inference in computationally intensive models, such as those based on numerical simulations. VISA extends importance-weighted forward-KL variational inference by employing a sequence of sample-average approximations, which are considered valid inside a trust region. This makes it possible… ▽ More We present variational inference with sequential sample-average approximation (VISA), a method for approximate inference in computationally intensive models, such as those based on numerical simulations. VISA extends importance-weighted forward-KL variational inference by employing a sequence of sample-average approximations, which are considered valid inside a trust region. This makes it possible to reuse model evaluations across multiple gradient steps, thereby reducing computational cost. We perform experiments on high-dimensional Gaussians, Lotka-Volterra dynamics, and a Pickover attractor, which demonstrate that VISA can achieve comparable approximation accuracy to standard importance-weighted forward-KL variational inference with computational savings of a factor two or more for conservatively chosen learning rates. △ Less

Submitted 15 March, 2024; v1 submitted 14 March, 2024; originally announced March 2024.

Topological Obstructions and How to Avoid Them

Authors: Babak Esmaeili, Robin Walters, Heiko Zimmermann, Jan-Willem van de Meent

Abstract: Incorporating geometric inductive biases into models can aid interpretability and generalization, but encoding to a specific geometric structure can be challenging due to the imposed topological constraints. In this paper, we theoretically and empirically characterize obstructions to training encoders with geometric latent spaces. We show that local optima can arise due to singularities (e.g. self… ▽ More Incorporating geometric inductive biases into models can aid interpretability and generalization, but encoding to a specific geometric structure can be challenging due to the imposed topological constraints. In this paper, we theoretically and empirically characterize obstructions to training encoders with geometric latent spaces. We show that local optima can arise due to singularities (e.g. self-intersection) or due to an incorrect degree or winding number. We then discuss how normalizing flows can potentially circumvent these obstructions by defining multimodal variational distributions. Inspired by this observation, we propose a new flow-based model that maps data points to multimodal distributions over geometric spaces and empirically evaluate our model on 2 domains. We observe improved stability during training and a higher chance of converging to a homeomorphic encoder. △ Less

Submitted 12 December, 2023; originally announced December 2023.

Coronal Heating as Determined by the Solar Flare Frequency Distribution Obtained by Aggregating Case Studies

Authors: James Paul Mason, Alexandra Werth, Colin G. West, Allison A. Youngblood, Donald L. Woodraska, Courtney Peck, Kevin Lacjak, Florian G. Frick, Moutamen Gabir, Reema A. Alsinan, Thomas Jacobsen, Mohammad Alrubaie, Kayla M. Chizmar, Benjamin P. Lau, Lizbeth Montoya Dominguez, David Price, Dylan R. Butler, Connor J. Biron, Nikita Feoktistov, Kai Dewey, N. E. Loomis, Michal Bodzianowski, Connor Kuybus, Henry Dietrick, Aubrey M. Wolfe , et al. (977 additional authors not shown)

Abstract: Flare frequency distributions represent a key approach to addressing one of the largest problems in solar and stellar physics: determining the mechanism that counter-intuitively heats coronae to temperatures that are orders of magnitude hotter than the corresponding photospheres. It is widely accepted that the magnetic field is responsible for the heating, but there are two competing mechanisms th… ▽ More Flare frequency distributions represent a key approach to addressing one of the largest problems in solar and stellar physics: determining the mechanism that counter-intuitively heats coronae to temperatures that are orders of magnitude hotter than the corresponding photospheres. It is widely accepted that the magnetic field is responsible for the heating, but there are two competing mechanisms that could explain it: nanoflares or Alfvén waves. To date, neither can be directly observed. Nanoflares are, by definition, extremely small, but their aggregate energy release could represent a substantial heating mechanism, presuming they are sufficiently abundant. One way to test this presumption is via the flare frequency distribution, which describes how often flares of various energies occur. If the slope of the power law fitting the flare frequency distribution is above a critical threshold, $α=2$ as established in prior literature, then there should be a sufficient abundance of nanoflares to explain coronal heating. We performed $>$600 case studies of solar flares, made possible by an unprecedented number of data analysts via three semesters of an undergraduate physics laboratory course. This allowed us to include two crucial, but nontrivial, analysis methods: pre-flare baseline subtraction and computation of the flare energy, which requires determining flare start and stop times. We aggregated the results of these analyses into a statistical study to determine that $α= 1.63 \pm 0.03$. This is below the critical threshold, suggesting that Alfvén waves are an important driver of coronal heating. △ Less

Submitted 9 May, 2023; originally announced May 2023.

Comments: 1,002 authors, 14 pages, 4 figures, 3 tables, published by The Astrophysical Journal on 2023-05-09, volume 948, page 71

A Variational Perspective on Generative Flow Networks

Authors: Heiko Zimmermann, Fredrik Lindsten, Jan-Willem van de Meent, Christian A. Naesseth

Abstract: Generative flow networks (GFNs) are a class of models for sequential sampling of composite objects, which approximate a target distribution that is defined in terms of an energy function or a reward. GFNs are typically trained using a flow matching or trajectory balance objective, which matches forward and backward transition models over trajectories. In this work, we define variational objectives… ▽ More Generative flow networks (GFNs) are a class of models for sequential sampling of composite objects, which approximate a target distribution that is defined in terms of an energy function or a reward. GFNs are typically trained using a flow matching or trajectory balance objective, which matches forward and backward transition models over trajectories. In this work, we define variational objectives for GFNs in terms of the Kullback-Leibler (KL) divergences between the forward and backward distribution. We show that variational inference in GFNs is equivalent to minimizing the trajectory balance objective when sampling trajectories from the forward model. We generalize this approach by optimizing a convex combination of the reverse- and forward KL divergence. This insight suggests variational inference methods can serve as a means to define a more general family of objectives for training generative flow networks, for example by incorporating control variates, which are commonly used in variational inference, to reduce the variance of the gradients of the trajectory balance objective. We evaluate our findings and the performance of the proposed variational objective numerically by comparing it to the trajectory balance objective on two synthetic tasks. △ Less

Submitted 14 October, 2022; originally announced October 2022.

Nested Variational Inference

Authors: Heiko Zimmermann, Hao Wu, Babak Esmaeili, Jan-Willem van de Meent

Abstract: We develop nested variational inference (NVI), a family of methods that learn proposals for nested importance samplers by minimizing an forward or reverse KL divergence at each level of nesting. NVI is applicable to many commonly-used importance sampling strategies and provides a mechanism for learning intermediate densities, which can serve as heuristics to guide the sampler. Our experiments appl… ▽ More We develop nested variational inference (NVI), a family of methods that learn proposals for nested importance samplers by minimizing an forward or reverse KL divergence at each level of nesting. NVI is applicable to many commonly-used importance sampling strategies and provides a mechanism for learning intermediate densities, which can serve as heuristics to guide the sampler. Our experiments apply NVI to (a) sample from a multimodal distribution using a learned annealing path (b) learn heuristics that approximate the likelihood of future observations in a hidden Markov model and (c) to perform amortized inference in hierarchical deep generative models. We observe that optimizing nested objectives leads to improved sample quality in terms of log average weight and effective sample size. △ Less

Submitted 21 June, 2021; originally announced June 2021.

Learning Proposals for Probabilistic Programs with Inference Combinators

Authors: Sam Stites, Heiko Zimmermann, Hao Wu, Eli Sennesh, Jan-Willem van de Meent

Abstract: We develop operators for construction of proposals in probabilistic programs, which we refer to as inference combinators. Inference combinators define a grammar over importance samplers that compose primitive operations such as application of a transition kernel and importance resampling. Proposals in these samplers can be parameterized using neural networks, which in turn can be trained by optimi… ▽ More We develop operators for construction of proposals in probabilistic programs, which we refer to as inference combinators. Inference combinators define a grammar over importance samplers that compose primitive operations such as application of a transition kernel and importance resampling. Proposals in these samplers can be parameterized using neural networks, which in turn can be trained by optimizing variational objectives. The result is a framework for user-programmable variational methods that are correct by construction and can be tailored to specific models. We demonstrate the flexibility of this framework by implementing advanced variational methods based on amortized Gibbs sampling and annealing. △ Less

Submitted 16 June, 2021; v1 submitted 28 February, 2021; originally announced March 2021.

Comments: Accepted to UAI 2021

Hyperpolarization of cis-15N,15N'-azobenzene by parahydrogen at ultralow magnetic fields

Authors: Kirill F. Sheberstov, Vitaly P. Kozinenko, Alexey S. Kiryutin, Hans-Martin Vieth, Herbert Zimmermann, Konstantin L. Ivanov, Alexandra V. Yurkovskaya

Abstract: Development of the methods to exploit nuclear hyperpolarization and search for molecules whose nuclear spins can be efficiently hyperpolarized is an active area in nuclear magnetic resonance. Of particular interest are those molecules that have long nuclear relaxation times, making them to be suitable candidates as contrast agents in magnetic resonance imaging. In this work, we present a detailed… ▽ More Development of the methods to exploit nuclear hyperpolarization and search for molecules whose nuclear spins can be efficiently hyperpolarized is an active area in nuclear magnetic resonance. Of particular interest are those molecules that have long nuclear relaxation times, making them to be suitable candidates as contrast agents in magnetic resonance imaging. In this work, we present a detailed study of SABRE SHEATH (Signal Amplification By Reversible Exchange in Shield Enabled Alignment Transfer to Heteronuclei) experiments of 15N,15N' azobenzene. In SABRE SHEATH experiments nuclear spins of the target are hyperpolarized by transfer of spin polarization from parahydrogen at ultralow fields during a reversible chemical process. The studied system is complicated, and we are concerned only about a subset of the data, presenting details for the molecules that experience fast chemical exchange at the catalytic complex and thus are involved in polarizing the free azobenzene. Azobenzene exists in two isomers trans- and cis-. We show that all nuclear spins in cis-azobenzene can be efficiently hyperpolarized by SABRE at suitable magnetic fields. Enhancement factors (relative to 9.4 T) reach several thousands of times for 15N spins and a few tens of times for the 1H spins. There are two approaches to observe either hyperpolarized magnetization of 15N/1H spins or hyperpolarized singlet order of the 15N spin pair. We compare these approaches and present the field dependencies of SABRE experiments for them. No hyperpolarization of trans 15N,15N' azobenzene was observed. The results presented here will be useful for further experiments where hyperpolarized cis-15N,15N' azobenzene is switched by light to trans 15N,15N' azobenzene for storing the produced hyperpolarization in the long-lived spin state of the 15N pair of trans-15N,15N' azobenzene. △ Less

Submitted 27 February, 2021; originally announced March 2021.

On Error Correction Neural Networks for Economic Forecasting

Authors: Mhlasakululeka Mvubu, Emmanuel Kabuga, Christian Plitz, Bubacarr Bah, Ronnie Becker, Hans Georg Zimmermann

Abstract: Recurrent neural networks (RNNs) are more suitable for learning non-linear dependencies in dynamical systems from observed time series data. In practice all the external variables driving such systems are not known a priori, especially in economical forecasting. A class of RNNs called Error Correction Neural Networks (ECNNs) was designed to compensate for missing input variables. It does this by f… ▽ More Recurrent neural networks (RNNs) are more suitable for learning non-linear dependencies in dynamical systems from observed time series data. In practice all the external variables driving such systems are not known a priori, especially in economical forecasting. A class of RNNs called Error Correction Neural Networks (ECNNs) was designed to compensate for missing input variables. It does this by feeding back in the current step the error made in the previous step. The ECNN is implemented in Python by the computation of the appropriate gradients and it is tested on stock market predictions. As expected it out performed the simple RNN and LSTM and other hybrid models which involve a de-noising pre-processing step. The intuition for the latter is that de-noising may lead to loss of information. △ Less

Submitted 1 June, 2020; v1 submitted 10 April, 2020; originally announced April 2020.

Comments: 13 pages, 4 figures, 11 tables

MSC Class: 37M10; 62M10; 91B84 ACM Class: I.2.6; I.5.1; I.5.4

Amortized Population Gibbs Samplers with Neural Sufficient Statistics

Authors: Hao Wu, Heiko Zimmermann, Eli Sennesh, Tuan Anh Le, Jan-Willem van de Meent

Abstract: We develop amortized population Gibbs (APG) samplers, a class of scalable methods that frames structured variational inference as adaptive importance sampling. APG samplers construct high-dimensional proposals by iterating over updates to lower-dimensional blocks of variables. We train each conditional proposal by minimizing the inclusive KL divergence with respect to the conditional posterior. To… ▽ More We develop amortized population Gibbs (APG) samplers, a class of scalable methods that frames structured variational inference as adaptive importance sampling. APG samplers construct high-dimensional proposals by iterating over updates to lower-dimensional blocks of variables. We train each conditional proposal by minimizing the inclusive KL divergence with respect to the conditional posterior. To appropriately account for the size of the input data, we develop a new parameterization in terms of neural sufficient statistics. Experiments show that APG samplers can train highly structured deep generative models in an unsupervised manner, and achieve substantial improvements in inference accuracy relative to standard autoencoding variational methods. △ Less

Submitted 9 July, 2020; v1 submitted 4 November, 2019; originally announced November 2019.

Constant-adiabaticity RF-pulses for generating long-lived singlet spin states in NMR

Authors: Bogdan A. Rodin, Kirill F. Sheberstov, Alexey S. Kiryutin, Joseph T. Hill-Cousins, Lynda J. Brown, Richard C. D. Brown, Baptiste Jamain, Herbert Zimmermann, Renad Z. Sagdeev, Alexandra V. Yurkovskaya, Konstantin L. Ivanov

Abstract: A method is implemented to perform "fast" adiabatic variation of the spin Hamiltonian by imposing the constant adiabaticity condition. The method is applied to improve the performance of singlet-state Nuclear Magnetic Resonance (NMR) experiments, specifically, for efficient generation and readout of the singlet spin order in coupled spin pairs by applying adiabatically ramped RF-fields. Test exper… ▽ More A method is implemented to perform "fast" adiabatic variation of the spin Hamiltonian by imposing the constant adiabaticity condition. The method is applied to improve the performance of singlet-state Nuclear Magnetic Resonance (NMR) experiments, specifically, for efficient generation and readout of the singlet spin order in coupled spin pairs by applying adiabatically ramped RF-fields. Test experiments have been performed on a specially designed molecule having two strongly coupled C-13 spins and on selectively isotopically labelled glycerol having two pairs of coupled protons. Optimized RF-ramps show improved performance in comparison, for example, to linear ramps. We expect that the methods described here are useful, not only for singlet-state NMR experiments, but also for other experiments in magnetic resonance, which utilize adiabatic variation of the spin Hamiltonian. △ Less

Submitted 30 October, 2018; originally announced October 2018.

Comments: 14 pages, 13 figures, 2 tables

Scintillation light produced by low-energy beams of highly-charged ions

Authors: M. Vogel, D. F. A. Winters, H. Ernst, H. Zimmermann, O. Kester

Abstract: Measurements have been performed of scintillation light intensities emitted from various inorganic scintillators irradiated with low-energy beams of highly-charged ions from an electron beam ion source (EBIS) and an electron cyclotron resonance ion source (ECRIS). Beams of xenon ions Xe$^{q+}$ with various charge states between $q$=2 and $q$=18 have been used at energies between 5 keV and 17.5 k… ▽ More Measurements have been performed of scintillation light intensities emitted from various inorganic scintillators irradiated with low-energy beams of highly-charged ions from an electron beam ion source (EBIS) and an electron cyclotron resonance ion source (ECRIS). Beams of xenon ions Xe$^{q+}$ with various charge states between $q$=2 and $q$=18 have been used at energies between 5 keV and 17.5 keV per charge generated by the ECRIS. The intensity of the beam was typically varied between 1 and 100 nA. Beams of highly charged residual gas ions have been produced by the EBIS at 4.5 keV per charge and with low intensities down to 100 pA. The scintillator materials used are flat screens of P46 YAG and P43 phosphor. In all cases, scintillation light emitted from the screen surface was detected by a CCD camera. The scintillation light intensity has been found to depend linearly on the kinetic ion energy per time deposited into the scintillator, while up to $q$=18 no significant contribution from the ions' potential energy was found. We discuss the results on the background of a possible use as beam diagnostics e.g. for the new HITRAP facility at GSI, Germany. △ Less

Submitted 25 October, 2007; originally announced October 2007.

Comments: 6 pages, 8 figures

Journal ref: Nucl.Instrum.Meth.B263:518-522,2007

Relaxation in the glass-former acetyl salicylic acid studied by deuteron magnetic resonance and dielectric spectroscopy

Authors: R. Nath, T. El Goresy, H. Zimmermann, R. Bohmer

Abstract: Supercooled liquid and glassy acetyl salicylic acid was studied using dielectric spectroscopy and deuteron relaxometry in a wide temperature range. The supercooled liquid is characterized by major deviations from thermally activated behavior. In the glass the secondary relaxation exhibits the typical features of a Johari-Goldstein process. Via measurements of spin-lattice relaxation times the se… ▽ More Supercooled liquid and glassy acetyl salicylic acid was studied using dielectric spectroscopy and deuteron relaxometry in a wide temperature range. The supercooled liquid is characterized by major deviations from thermally activated behavior. In the glass the secondary relaxation exhibits the typical features of a Johari-Goldstein process. Via measurements of spin-lattice relaxation times the selectively deuterated methyl group was used as a sensitive probe of its local environments. There is a large difference in the mean activation energy in the glass with respect to that in crystalline acetyl salicylic acid. This can be understood by taking into account the broad energy barrier distribution in the glass. △ Less

Submitted 9 April, 2006; originally announced April 2006.

Comments: 8 pages, 3 figures, Submitted to Phys. Rev. E

Journal ref: Phys. Rev. E 74, 021506 (2006)

XMM-Newton observation of SN1993J in M81

Authors: H. -U. Zimmermann, B. Aschenbach

Abstract: In April 2001 SN1993J was observed with both the PN and MOS cameras of the XMM-Newton observatory, resulting in about 7. x 10^4 s of acceptable observation time. Fit results with both the PN and MOS2 camera spectra studying different spectral models are presented. The spectra are best fitted in the energy range between 0.3 and 11 keV by a 2-component thermal model with temperatures of kT_1 = 0.3… ▽ More In April 2001 SN1993J was observed with both the PN and MOS cameras of the XMM-Newton observatory, resulting in about 7. x 10^4 s of acceptable observation time. Fit results with both the PN and MOS2 camera spectra studying different spectral models are presented. The spectra are best fitted in the energy range between 0.3 and 11 keV by a 2-component thermal model with temperatures of kT_1 = 0.34+-0.04 keV and kT_2 = 6.54+-4 keV, adopting ionization equilibrium. A fit with a shock model also provides acceptable results. Combining the XMM-Newton data with former X-ray observations of the supernova, we discuss the general trend of L_x propto t^{-0.30} and the bump of the X-ray light curve as well as former and recent spectral results in the light of the standard SN model as first proposed by Chevalier in 1982. △ Less

Submitted 17 April, 2003; originally announced April 2003.

Comments: 7 pages, 3 figures

Journal ref: Astron.Astrophys. 406 (2003) 969-974

Funneling with the Two-Beam RFQ

Authors: H. Zimmermann

Abstract: New high current accelerator facilities like proposed for HIDIF or ESS require a beam with a high brilliance. These beams can not be produced by a single pass rf-linac. The increase in brightness in such a driver linac is done by several funneling stages at low energies, in which two identically bunched ion beams are combined into a single beam with twice the frequency current and brightness. Ou… ▽ More New high current accelerator facilities like proposed for HIDIF or ESS require a beam with a high brilliance. These beams can not be produced by a single pass rf-linac. The increase in brightness in such a driver linac is done by several funneling stages at low energies, in which two identically bunched ion beams are combined into a single beam with twice the frequency current and brightness. Our Two-Beam-RFQ funneling experiment is a setup of two ion sources, a two beam RFQ, a funnel deflector and beam diagnostic equipment to demonstrate funneling of ion beams as a model for the first funneling stage of a HIIF driver. The progress of the funneling experiment and results of simulations will be presented. △ Less

Submitted 14 August, 2000; originally announced August 2000.

Comments: 3 pages, 11 figures

Report number: IAP03082000

Journal ref: eConf C000821 (2000) THB02

The X-ray Spectrum and Light Curve of Supernova 1995N

Authors: D. W. Fox, W. H. G. Lewin, A. Fabian, K. Iwasawa, R. Terlevich, H. U. Zimmermann, B. Aschenbach, K. Weiler, S. Van Dyk, R. Chevalier, R. Rutledge, H. Inoue, S. Uno

Abstract: We report on multi-epoch X-ray observations of the Type IIn (narrow emission line) supernova SN 1995N with the ROSAT and ASCA satellites. The January 1998 ASCA X-ray spectrum is well fitted by a thermal bremsstrahlung (kT~10 keV, N_H~6e20 cm^-2) or power-law (alpha~1.7, N_H~1e21 cm^-2) model. The X-ray light curve shows evidence for significant flux evolution between August 1996 and January 1998… ▽ More We report on multi-epoch X-ray observations of the Type IIn (narrow emission line) supernova SN 1995N with the ROSAT and ASCA satellites. The January 1998 ASCA X-ray spectrum is well fitted by a thermal bremsstrahlung (kT~10 keV, N_H~6e20 cm^-2) or power-law (alpha~1.7, N_H~1e21 cm^-2) model. The X-ray light curve shows evidence for significant flux evolution between August 1996 and January 1998: the count rate from the source decreased by 30% between our August 1996 and August 1997 ROSAT observations, and the X-ray luminosity most likely increased by a factor of ~2 between our August 1997 ROSAT and January 1998 ASCA observations, although evolution of the spectral shape over this interval is not ruled out. The high X-ray luminosity, L_X~1e41 erg/sec, places SN 1995N in a small group of Type IIn supernovae with strong circumstellar interaction, and the evolving X-ray luminosity suggests that the circumstellar medium is distributed inhomogeneously. △ Less

Submitted 25 August, 2000; originally announced August 2000.

Comments: MNRAS accepted. 6 pages, 2 figures; uses mn.sty and psfig

Report number: CSR-00-54

Journal ref: Mon.Not.Roy.Astron.Soc.319:1154,2000

Vibrational States of Glassy and Crystalline Orthotherphenyl

Authors: A. Toelle, H. Zimmermann, F. Fujara, W. Petry, W. Schmidt, H. Schober, J. Wuttke

Abstract: Low-frequency vibrations of glassy and crystalline orthoterphenyl are studied by means of neutron scattering. Phonon dispersions are measured along the main axes of a single crystal, and the corresponding longitudinal and transversal sound velocities are obtained. For glassy and polycrystalline samples, a density of vibrational states is determined and cross-checked against other dynamic observa… ▽ More Low-frequency vibrations of glassy and crystalline orthoterphenyl are studied by means of neutron scattering. Phonon dispersions are measured along the main axes of a single crystal, and the corresponding longitudinal and transversal sound velocities are obtained. For glassy and polycrystalline samples, a density of vibrational states is determined and cross-checked against other dynamic observables. In the crystal, low-lying zone-boundary modes lead to an excess over the Debye density of states. In the glass, the boson peak is located at even lower frequencies. With increasing temperature, both glass and crystal show anharmonicity. △ Less

Submitted 18 February, 2000; originally announced February 2000.

Comments: 7 pages of LaTeX (svjour), 2 tables, 10 figures accepted in Eur. Phys. J. B

Report number: EIIIb-at-0001

Journal ref: Eur. Phys. J. B 16, 73 (2000)

The ROSAT All-Sky Survey Bright Source Catalogue

Authors: Wolfgang Voges, Bernd Aschenbach, Thomas Boller, Heinrich Braeuninger, Ulrich Briel, Wolfgang Burkert, Konrad Dennerl, Jakob Englhauser, Rainer Gruber, Frank Haberl, Gisela Hartner, Guenther Hasinger, Elmar Pfeffermann, Wolfgang Pietsch, Peter Predehl, Cristina Rosso, Juergen H. M. M. Schmitt, Joachim Truemper, Hans-Ulrich Zimmermann

Abstract: We present the ROSAT All-Sky Survey Bright Source Catalogue (RASS-BSC, revision 1RXS) derived from the all-sky survey performed during the first half year (1990/91) of the ROSAT mission. 18,811 sources are catalogued (i) down to a limiting ROSAT PSPC count-rate of 0.05 cts/s in the 0.1-2.4 keV energy band, (ii) with a detection likelihood of at least 15 and (iii) at least 15 source counts. The 1… ▽ More We present the ROSAT All-Sky Survey Bright Source Catalogue (RASS-BSC, revision 1RXS) derived from the all-sky survey performed during the first half year (1990/91) of the ROSAT mission. 18,811 sources are catalogued (i) down to a limiting ROSAT PSPC count-rate of 0.05 cts/s in the 0.1-2.4 keV energy band, (ii) with a detection likelihood of at least 15 and (iii) at least 15 source counts. The 18,811 sources underwent both an automatic validation and an interactive visual verification process in which for 94% of the sources the results of the standard processing were confirmed. The remaining 6% have been analyzed using interactive methods and these sources have been flagged. Flags are given for (i) nearby sources; (ii) sources with positional errors; (iii) extended sources; (iv) sources showing complex emission structures; and (v) sources which are missed by the standard analysis software. Broad band (0.1-2.4 keV) images are available for sources flagged by (ii), (iii) and (iv). For each source the ROSAT name, position in equatorial coordinates, positional error, source count-rate and error, background count-rate, exposure time, two hardness-ratios and errors, extent and likelihood of extent, likelihood of detection, and the source extraction radius are provided. At a brightness limit of 0.1 cts/s (8,547 sources) the catalogue represents a sky coverage of 92%. The RASS-BSC, the table of possible identification candidates, and the broad band images are available in electronic form (Voges et al. 1996a) via http://wave.xray.mpe.mpg.de/rosat/catalogues/rass-bsc . △ Less

Submitted 20 September, 1999; v1 submitted 17 September, 1999; originally announced September 1999.

Comments: 17 pages, 18 figures, corrected fig. 3b, to appear in A&A 349, 389-405 (1999), version with colour plots available from http://wave.xray.mpe.mpg.de/rosat/catalogues/rass-bsc/pub/rbsc-paper.ps.gz

Journal ref: Astron.Astrophys. 349 (1999) 389

Properties of new X-ray selected AGN

Authors: K. Bischoff, W. Pietsch, T. Boller, S. Döbereiner, W. Kollatschny, H. -U. Zimmermann

Abstract: We present the results of a program to identify so far unknown active nuclei (AGN) in galaxies. Candidate galactic nuclei have been selected for optical spectroscopy from a cross-correlation of the ROSAT all sky survey (RASS) bright source catalog with optical galaxy catalogs. In this paper we present results from the second optical observing run. We present the results of a program to identify so far unknown active nuclei (AGN) in galaxies. Candidate galactic nuclei have been selected for optical spectroscopy from a cross-correlation of the ROSAT all sky survey (RASS) bright source catalog with optical galaxy catalogs. In this paper we present results from the second optical observing run. △ Less

Submitted 7 October, 1998; originally announced October 1998.

Comments: 4 pages LaTeX (modified aa style using pxray_symp.cls), incl. 5 Postscript figures, to appear in: B. Aschenbach et al. (eds.) Highlights in X-ray Astronomy (Garching, June 1998)

Report number: USWKB-98-02

New active galactic nuclei detected in ROSAT All Sky Survey galaxies; Part I: Verification of selection strategy

Authors: W. Pietsch, Th. Boller, S. Doebereiner, H. -U. Zimmermann, K. Bischoff, W. Kollatschny

Abstract: We present the first results of a program to identify so far unknown active nuclei (AGN) in galaxies. Candidate galactic nuclei have been selected for optical spectroscopy from a cross-correlation of the ROSAT All Sky Survey (RASS) bright source catalog with optical galaxy catalogs. A high X-ray flux has been used as pointer to galaxies with a high probability to contain active nuclei. Only gala… ▽ More We present the first results of a program to identify so far unknown active nuclei (AGN) in galaxies. Candidate galactic nuclei have been selected for optical spectroscopy from a cross-correlation of the ROSAT All Sky Survey (RASS) bright source catalog with optical galaxy catalogs. A high X-ray flux has been used as pointer to galaxies with a high probability to contain active nuclei. Only galaxies have been accepted for the program for which no activity was noted in NED. For many of the galaxies no radial velocity was reported before. The optical spectra demonstrate that the galaxies cover a redshift range of 0.014 to 0.13 and that most of them host active nuclei. For 75 percent of the 33 candidates the X-ray emission is caused by the AGN. In addition several of the remaining candidates host Seyfert 2/LINER nuclei that, however, most certainly cannot explain the X-ray emission alone. Three BL Lac objects have been detected serendipitously in galaxy fields that have been followed up by short ROSAT HRI observations to confirm the X-ray galaxy identification with improved position accuracy and point response function. The sources show X-ray to radio flux ratios typical for X-ray selected BL Lac objects. The results presented in the paper prove the selection strategy as very successful to detect previously unknown AGN of all Seyfert 1 types in nearby galaxies encouraging the extension of this program. The detection of new nearby AGN will be used to initiate a detailed investigation of their multi-wavelength properties and a comparison with the more distant AGN population. △ Less

Submitted 22 January, 1998; originally announced January 1998.

Comments: 17 pages incl. 5 figures, accepted for publication in Astronomy and Astrophysics, full paper version available at http://www.rosat.mpe-garching.mpg.de/~wnp/