Balázs Kégl

Balázs Kégl
French National Centre for Scientific Research | CNRS · Linear Accelerator Laboratory

About

174
Publications
33,808
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
11,787
Citations
Citations since 2016
0 Research Items
6733 Citations
201620172018201920202021202202004006008001,0001,200
201620172018201920202021202202004006008001,0001,200
201620172018201920202021202202004006008001,0001,200
201620172018201920202021202202004006008001,0001,200

Publications

Publications (174)
Article
We present a detailed statistical treatment of the energy calibration of hybrid air-shower detectors, which combine a surface detector array and a fluorescence detector, to obtain an unbiased estimate of the calibration curve. The special features of calibration data from air showers prevent unbiased results, if a standard least-squares fit is appl...
Article
14 pages, 5 figures, submitted to Phys. Rev. D - See paper for full list of authors
Article
22 pages, 9 figures, submitted to Journal of Cosmology and Astroparticle Physics (JCAP) - See paper for full list of authors
Article
See paper for full list of authors – 12 pages, 6 figures, accepted by Physical Review D
Article
See paper for full list of authors – Paper submitted to NIM A
Article
Full-text available
Understanding how evolution of antimicrobial resistance increases resistance to other drugs is a challenge of profound importance. By combining experimental evolution and genome sequencing of 63 laboratory-evolved lines, we charted a map of cross-resistance interactions between antibiotics in Escherichia coli, and explored the driving evolutionary...
Article
See paper for full list of authors - Accepted for publication in Phys. Rev. D; 50 pages, 15 figures
Article
17 pages, accepted for publication in The Astrophysical Journal Letters ; see paper for full list of authors
Article
15 pages, 9 figures, accepted for publication in Physical Review D ; see paper for full list of authors
Article
28 pages, 10 figures, accepted for publication in The Astrophysical Journal
Article
Full-text available
Within the framework of AdaBoost.MH, we propose to train vector-valued decision trees to optimize the multi-class edge without reducing the multi-class problem to $K$ binary one-against-all classifications. The key element of the method is a vector-valued decision stump, factorized into an input-independent vector of length $K$ and label-independen...
Article
Full-text available
Motivated by an abstract notion of low-level edge detector filters, we propose a simple method of unsupervised feature construction based on pairwise statistics of features. In the first step, we construct neighborhoods of features by regrouping features that correlate. Then we use these subsets as filters to produce new neighborhood features. Next...
Article
Full-text available
In subset ranking, the goal is to learn a ranking function that approximates a gold standard partial ordering of a set of objects (in our case, a set of documents retrieved for the same query). The partial ordering is given by relevance labels representing the relevance of documents with respect to the query on an absolute scale. Our approach consi...
Article
Full-text available
The evolution of resistance to a single antibiotic is frequently accompanied by increased resistance to multiple other antimicrobial agents. In sharp contrast, very little is known about the frequency and mechanisms underlying collateral sensitivity. In this case, genetic adaptation under antibiotic stress yields enhanced sensitivity to other antib...
Article
Full-text available
The Pierre Auger Observatory is the world's largest cosmic ray observatory. Our current exposure reaches nearly 40,000 km$^2$ str and provides us with an unprecedented quality data set. The performance and stability of the detectors and their enhancements are described. Data analyses have led to a number of major breakthroughs. Among these we discu...
Article
Full-text available
Joint contributions of the Pierre Auger and Telescope Array Collaborations to the 33rd International Cosmic Ray Conference, Rio de Janeiro, Brazil, July 2013: cross-calibration of the fluorescence telescopes, large scale anisotropies and mass composition.
Article
Full-text available
Contributions of the Pierre Auger Collaboration to the 33rd International Cosmic Ray Conference, Rio de Janeiro, Brazil, July 2013
Article
Full-text available
Multivariate discrimination or classification is one of the best-studied problem in machine learning, with a plethora of well-tested and well-performing algorithms. There are also several good general textbooks [1-9] on the subject written to an average engineering, computer science, or statistics graduate student; most of them are also accessible...
Conference Paper
Full-text available
The multi-armed bandit problem has attracted remarkable attention in the machine learning community and many efficient algorithms have been proposed to handle the so-called exploitation-exploration dilemma in various bandit setups. At the same time, significantly less effort has been devoted to adapting bandit algorithms to particular architectures...
Article
Hyperparameter learning has traditionally been a manual task because of the limited number of trials. Today's computing infrastructures allow bigger evaluation budgets, thus opening the way for algorithmic approaches. Recently, surrogate-based optimization was successfully applied to hyperparameter learning for deep belief networks and to WEKA clas...
Article
Full-text available
Grids organize resource sharing, a fundamental requirement of large scientific collaborations. Seamless integration of grids into everyday use requires responsiveness, which can be provided by elastic Clouds, in the Infrastructure as a Service (IaaS) paradigm. This paper proposes a model-free resource provisioning strategy supporting both requireme...
Article
Full-text available
When targeting a distribution that is artificially invariant under some permutations, Markov chain Monte Carlo (MCMC) algorithms face the label-switching problem, rendering marginal inference particularly cumbersome. Such a situation arises, for example, in the Bayesian analysis of finite mixture models. Adaptive MCMC algorithms such as adaptive Me...
Article
The Pierre Auger Observatory is exploring the potential of the radio detection technique to study extensive air showers induced by ultra-high energy cosmic rays. The Auger Engineering Radio Array (AERA) addresses both technological and scientific aspects of the radio technique. A first phase of AERA has been operating since September 2010 with dete...
Conference Paper
Full-text available
We focus on the problem of data mining over large-scale fully distributed databases, where each node stores only one data record. We assume that a data record is never allowed to leave the node it is stored at. Possible motivations for this assumption include privacy or a lack of a centralized infrastructure. To tackle this problem, earlier we prop...
Article
Full-text available
The Pierre Auger Observatory is a facility built to detect air showers produced by cosmic rays above 10^17 eV. During clear nights with a low illuminated moon fraction, the UV fluorescence light produced by air showers is recorded by optical telescopes at the Observatory. To correct the observations for variations in atmospheric conditions, atmosph...
Article
Full-text available
The surface detector array of the Pierre Auger Observatory can detect neutrinos with energy E ν between 10 17 eV and 10 20 eV from point-like sources across the sky south of +55° and north of –65° declinations. A search has been performed for highly inclined extensive air showers produced by the interaction of neutrinos of all flavors in the atmosp...
Article
Full-text available
In this paper we propose an algorithm that builds sparse decision DAGs (directed acyclic graphs) from a list of base classifiers provided by an external learning method such as AdaBoost. The basic idea is to cast the DAG design task as a Markov decision process. Each instance can decide to use or to skip each base classifier, based on the current s...
Article
Adaptive Metropolis (AM) is a powerful recent algorithmic tool in numerical Bayesian data analysis. AM builds on a well-known Markov Chain Monte Carlo algorithm but optimizes the rate of convergence to the target distribution by automatically tuning the design parameters of the algorithm on the fly. Label switching is a major problem in inference o...
Article
Full-text available
The Surface Detector of the Pierre Auger Observatory is sensitive to neutrinos of all flavours above 0.1 EeV. These interact through charged and neutral currents in the atmosphere giving rise to extensive air showers. When interacting deeply in the atmosphere at nearly horizontal incidence, neutrinos can be distinguished from regular hadronic cosmi...
Article
Atmospheric conditions at the site of a cosmic ray observatory must be known for reconstructing observed extensive air showers. The Global Data Assimilation System (GDAS) is a global atmospheric model predicated on meteorological measurements and numerical weather predictions. GDAS provides altitude-dependent profiles of the main state variables of...
Article
Full-text available
The Surface Detector of the Pierre Auger Observatory is sensitive to neutrinos of all flavors above 0.1 EeV. These interact through charged and neutral currents in the atmosphere giving rise to extensive air showers. When interacting deeply in the atmosphere at nearly horizontal incidence, neutrinos can be distinguished from regular hadronic cosmic...
Conference Paper
Full-text available
Several recent advances to the state of the art in image classification benchmarks have come from better configurations of existing techniques rather than novel ap-proaches to feature learning. Traditionally, hyper-parameter optimization has been the job of humans because they can be very efficient in regimes where only a few trials are possible. P...
Conference Paper
Full-text available
In this paper we propose an algorithm that builds sparse decision DAGs (directed acyclic graphs) out of a list of features or base classifiers. The basic idea is to cast the DAG design task as a Markov decision process. Each instance can decide to use or to skip each base classifier, based on the current state of the classifier being built. The res...
Article
Full-text available
We present a comprehensive study of the influence of the geomagnetic field on the energy estimation of extensive air showers with a zenith angle smaller than $60^\circ$, detected at the Pierre Auger Observatory. The geomagnetic field induces an azimuthal modulation of the estimated energy of cosmic rays up to the ~2% level at large zenith angles. W...
Article
Full-text available
The surface detector array of the Pierre Auger Observatory consists of 1600 water-Cherenkov detectors, for the study of extensive air showers (EAS) generated by ultra-high-energy cosmic rays. We describe the trigger hierarchy, from the identification of candidate showers at the level of a single detector, amongst a large background (mainly random s...
Article
In this paper we introduce the concept of Lateral Trigger Probability (LTP) function, i.e., the probability for an extensive air shower (EAS) to trigger an individual detector of a ground based array as a function of distance to the shower axis, taking into account energy, mass and direction of the primary cosmic ray. We apply this concept to the s...
Article
We present the results of an analysis of data recorded at the Pierre Auger Observatory in which we search for groups of directionally-aligned events (or `multiplets') which exhibit a correlation between arrival direction and the inverse of the energy. These signatures are expected from sets of events coming from the same source after having been de...
Conference Paper
Full-text available
In subset ranking, the goal is to learn a ranking function that approximates a gold standard partial ordering of a set of objects (in our case, relevance labels of a set of documents retrieved for the same query). In this paper we introduce a learning to rank approach to subset ranking based on multi-class classification. Our technique can be summa...
Article
Contributions to 32nd International Cosmic Ray Conference, Beijing, China, August 2011 ; see paper for full list of authors
Article
Full-text available
Studies of the composition of the highest energy cosmic rays with the Pierre Auger Observatory, including examination of hadronic physics effects on the structure of extensive air showers.
Article
Full-text available
Astrophysical observations of ultra-high-energy cosmic rays with the Pierre Auger Observatory
Article
Full-text available
Technical reports on operations and monitoring of the Pierre Auger Observatory
Article
Full-text available
Ongoing and planned enhancements of the Pierre Auger Observatory
Article
Full-text available
Studies of the cosmic ray energy spectrum at the highest energies with the Pierre Auger Observatory
Article
Full-text available
Technical reports on operations and monitoring of the Pierre Auger Observatory
Article
Full-text available
The Pierre Auger Collaboration has reported evidence for anisotropy in the distribution of arrival directions of the cosmic rays with energies $E>E_{th}=5.5\times 10^{19}$ eV. These show a correlation with the distribution of nearby extragalactic objects, including an apparent excess around the direction of Centaurus A. If the particles responsible...
Article
Full-text available
Résumé : Two recurrent questions often appear when solving numerous real world policy search problems. First, the variables defining the so called Markov Decision Process are often continuous, that leads to the necessity for discretization of the considered state/action space or the use of a regression model, often non-linear, to approach the Q-fun...
Article
Full-text available
We present the results of searches for dipolar-type anisotropies in different energy ranges above $2.5\times 10^{17}$ eV with the surface detector array of the Pierre Auger Observatory, reporting on both the phase and the amplitude measurements of the first harmonic modulation in the right-ascension distribution. Upper limits on the amplitudes are...
Article
We present the results of searches for dipolar-type anisotropies in different energy ranges above 2.5 × 10^17 eV with the surface detector array of the Pierre Auger Observatory, reporting on both the phase and the amplitude measurements of the first harmonic modulation in the right-ascension distribution. Upper limits on the amplitudes are obtained...
Article
The Pierre Auger Observatory is a detector for ultra-high energy cosmic rays. It consists of a surface array to measure secondary particles at ground level and a fluorescence detector to measure the development of air showers in the atmosphere above the array. The “hybrid” detection mode combines the information from the two subsystems. We describe...
Article
The advent of the Auger Engineering Radio Array (AERA) necessitates the development of a powerful framework for the analysis of radio measurements of cosmic ray air showers. As AERA performs "radio-hybrid" measurements of air shower radio emission in coincidence with the surface particle detectors and fluorescence telescopes of the Pierre Auger Obs...