University of Sannio

About: University of Sannio is a education organization based out in Benevento, Italy. It is known for research contribution in the topics: Gravitational wave & LIGO. The organization has 1278 authors who have published 6125 publications receiving 167577 citations. The organization is also known as: Università degli Studi del Sannio & Universita degli Studi del Sannio.

Upper limits on the isotropic gravitational-wave background from Advanced LIGO and Advanced Virgo’s third observing run

Abstract: We report results of a search for an isotropic gravitational-wave background (GWB) using data from Advanced LIGO’s and Advanced Virgo’s third observing run (O3) combined with upper limits from the earlier O1 and O2 runs. Unlike in previous observing runs in the advanced detector era, we include Virgo in the search for the GWB. The results of the search are consistent with uncorrelated noise, and therefore we place upper limits on the strength of the GWB. We find that the dimensionless energy density Ω GW ≤ 5.8 × 10 − 9 at the 95% credible level for a flat (frequency-independent) GWB, using a prior which is uniform in the log of the strength of the GWB, with 99% of the sensitivity coming from the band 20–76.6 Hz; Ω GW ( f ) ≤ 3.4 × 10 − 9 at 25 Hz for a power-law GWB with a spectral index of 2 / 3 (consistent with expectations for compact binary coalescences), in the band 20–90.6 Hz; and Ω GW ( f ) ≤ 3.9 × 10 − 10 at 25 Hz for a spectral index of 3, in the band 20–291.6 Hz. These upper limits improve over our previous results by a factor of 6.0 for a flat GWB, 8.8 for a spectral index of 2 / 3 , and 13.1 for a spectral index of 3. We also search for a GWB arising from scalar and vector modes, which are predicted by alternative theories of gravity; we do not find evidence of these, and place upper limits on the strength of GWBs with these polarizations. We demonstrate that there is no evidence of correlated noise of magnetic origin by performing a Bayesian analysis that allows for the presence of both a GWB and an effective magnetic background arising from geophysical Schumann resonances. We compare our upper limits to a fiducial model for the GWB from the merger of compact binaries, updating the model to use the most recent data-driven population inference from the systems detected during O3a. Finally, we combine our results with observations of individual mergers and show that, at design sensitivity, this joint approach may yield stronger constraints on the merger rate of binary black holes at z ≳ 2 than can be achieved with individually resolved mergers alone.

An Empirical Study on Learning Bug-Fixing Patches in the Wild via Neural Machine Translation

Abstract: Millions of open-source projects with numerous bug fixes are available in code repositories. This proliferation of software development histories can be leveraged to learn how to fix common programming bugs. To explore such a potential, we perform an empirical study to assess the feasibility of using Neural Machine Translation techniques for learning bug-fixing patches for real defects. First, we mine millions of bug-fixes from the change histories of projects hosted on GitHub, in order to extract meaningful examples of such bug-fixes. Next, we abstract the buggy and corresponding fixed code, and use them to train an Encoder-Decoder model able to translate buggy code into its fixed version. In our empirical investigation we found that such a model is able to fix thousands of unique buggy methods in the wild. Overall, this model is capable of predicting fixed patches generated by developers in 9-50% of the cases, depending on the number of candidate patches we allow it to generate. Also, the model is able to emulate a variety of different Abstract Syntax Tree operations and generate candidate patches in a split second.

Who is going to mentor newcomers in open source projects

Abstract: When newcomers join a software project, they need to be properly trained to understand the technical and organizational aspects of the project. Inadequate training could likely lead to project delay or failure.In this paper we propose an approach, named Yoda (Young and newcOmer Developer Assistant) aimed at identifying and recommending mentors in software projects by mining data from mailing lists and versioning systems. Candidate mentors are identified among experienced developers who actively interact with newcomers. Then, when a newcomer joins the project, Yoda recommends her a mentor that, among the available ones, has already discussed topics relevant for the newcomer.Yoda has been evaluated on software repositories of five open source projects. We have also surveyed some developers of these projects to understand whether mentoring was actually performed in their projects, and asked them to evaluate the mentoring relations Yoda identified. Results indicate that top committers are not always the most appropriate mentors, and show the potential usefulness of Yoda as a recommendation system to aid project managers in supporting newcomers joining a software project.