arXiv subjects

Nathan Cooper

Publications and source records attributed to Nathan Cooper.

At least 37 records · Page 2Linked to original sources

arXiv2023

Translating Video Recordings of Complex Mobile App UI Gestures into Replayable Scenarios

Screen recordings of mobile applications are easy to obtain and capture a wealth of information pertinent to software developers (e.g., bugs or feature requests), making them a popular mechanism for crowdsourced app feedback. Thus, these videos are becoming a common artifact that developers must manage. In light of unique mobile development constraints, including swift release cycles and rapidly evolving platforms, automated techniques for analyzing all types of rich software artifacts provide benefit to mobile developers. Unfortunately, automatically analyzing screen recordings presents serious challenges, due to their graphical nature, compared to other types of (textual) artifacts. To address these challenges, this paper introduces V2S+, an automated approach for translating video recordings of Android app usages into replayable scenarios. V2S+ is based primarily on computer vision techniques and adapts recent solutions for object detection and image classification to detect and classify user gestures captured in a video, and convert these into a replayable test scenario. Given that V2S+ takes a computer vision-based approach, it is applicable to both hybrid and native Android applications. We performed an extensive evaluation of V2S+ involving 243 videos depicting 4,028 GUI-based actions collected from users exercising features and reproducing bugs from a collection of over 90 popular native and hybrid Android apps. Our results illustrate that V2S+ can accurately replay scenarios from screen recordings, and is capable of reproducing $\approx$ 90.2% of sequential actions recorded in native application scenarios on physical devices, and $\approx$ 83% of sequential actions recorded in hybrid application scenarios on emulators, both with low overhead. A case study with three industrial partners illustrates the potential usefulness of V2S+ from the viewpoint of developers.

cs.SE↗

arXiv2022

Observation of collectivity enhanced magnetoassociation of $^6$Li in the quantum degenerate regime

The association process of Feshbach molecules is well described by a Landau-Zener (LZ) transition above the Fermi temperature, such that two-body physics dominates the dynamics. However, using $^6$Li atoms and the associated Feshbach resonance at $B_r=834.1$ G, we observe an enhancement of the atom-molecule coupling as the fermionic atoms reach degeneracy, demonstrating the importance of many-body coherence not captured by the conventional LZ model. In the experiment, we apply a linear association ramp ranging from adiabatic to non-equilibrium molecule association for various temperatures. We develop a theoretical model that explains the temperature dependence of the atom-molecule coupling. Furthermore, we characterize this dependence experimentally and extract the atom-molecule coupling coefficient as a function of temperature, finding qualitative agreement between our model and experimental results. In addition, we simulate the dynamics of molecular association during a nonlinear field ramp. We find that, in the non-equilibrium regime, molecular association efficiency can be enhanced by sweeping the magnetic field cubically with time. Accurate measurement of the atom-molecule coupling coefficient is important for both theoretical and experimental studies of molecular association and many-body collective dynamics.

physics.atom-ph↗

arXiv2022

High-performance, additively-manufactured atomic spectroscopy apparatus for portable quantum technologies

We demonstrate a miniaturised and highly robust system for performing Doppler-free spectroscopy on thermal atomic vapour for three frequencies as required for cold atom-based quantum technologies. The application of additive manufacturing techniques, together with efficient use of optical components, produce a compact, stable optical system, with a volume of 0.089 L and a weight of 120 g. The device occupies less than a tenth of the volume of, and is considerably lower cost than, conventional spectroscopic systems, but also offers excellent stability against environmental disturbances. We characterise the response of the system to changes in environmental temperature between 10 and 30$^{\circ}$C and exposure to vibrations between 0 - 2000 Hz, finding that the system can reliably perform spectroscopic measurements despite substantial vibrational noise and temperature changes. Our results show that 3D-printed optical systems are an excellent solution for portable quantum technologies.

physics.atom-ph↗

arXiv2022

Using Transfer Learning for Code-Related Tasks

Deep learning (DL) techniques have been used to support several code-related tasks such as code summarization and bug-fixing. In particular, pre-trained transformer models are on the rise, also thanks to the excellent results they achieved in Natural Language Processing (NLP) tasks. The basic idea behind these models is to first pre-train them on a generic dataset using a self-supervised task (e.g, filling masked words in sentences). Then, these models are fine-tuned to support specific tasks of interest (e.g, language translation). A single model can be fine-tuned to support multiple tasks, possibly exploiting the benefits of transfer learning. This means that knowledge acquired to solve a specific task (e.g, language translation) can be useful to boost performance on another task (e.g, sentiment classification). While the benefits of transfer learning have been widely studied in NLP, limited empirical evidence is available when it comes to code-related tasks. In this paper, we assess the performance of the Text-To-Text Transfer Transformer (T5) model in supporting four different code-related tasks: (i) automatic bug-fixing, (ii) injection of code mutants, (iii) generation of assert statements, and (iv) code summarization. We pay particular attention in studying the role played by pre-training and multi-task fine-tuning on the model's performance. We show that (i) the T5 can achieve better performance as compared to state-of-the-art baselines; and (ii) while pre-training helps the model, not all tasks benefit from a multi-task fine-tuning.

cs.SE↗

arXiv2022

Dual-frequency Doppler-free spectroscopy for simultaneous laser stabilisation in compact atomic physics experiments

Vapour cell spectroscopy is an essential technique in many fields; in particular, nearly all atom and ion trapping experiments rely on simultaneous spectroscopy of two atomic transitions, traditionally employing separate apparatus for each transition. Here, we demonstrate simultaneous spectroscopy on two atomic transitions using spatially-overlapped beams from two independent lasers, within a single spectroscopic apparatus. We show that, in addition to aiding compactness, this approach offers superior performance, leading to sharper spectroscopy peaks and stronger absorption signals. Doppler-free locking features become visible over a frequency range several hundred MHz wider than for standard saturated absorption spectroscopy. By exploring the full, 2D parameter space associated with dual-frequency spectroscopy, we reveal a lattice-like structure of sharp resonance features in 2D frequency space, which enhances experimental versatility by allowing laser frequency stabilisation anywhere within a wide manifold of locations in 2D frequency space. The process of simultaneous frequency stabilisation of two lasers is analysed in detail, revealing that it can be expected to produce significant improvements in laser stability when compared to conventional, single-frequency spectroscopy.

physics.atom-ph↗

arXiv2021

An Empirical Study on the Usage of Transformer Models for Code Completion

Code completion aims at speeding up code writing by predicting the next code token(s) the developer is likely to write. Works in this field focused on improving the accuracy of the generated predictions, with substantial leaps forward made possible by deep learning (DL) models. However, code completion techniques are mostly evaluated in the scenario of predicting the next token to type, with few exceptions pushing the boundaries to the prediction of an entire code statement. Thus, little is known about the performance of state-of-the-art code completion approaches in more challenging scenarios in which, for example, an entire code block must be generated. We present a large-scale study exploring the capabilities of state-of-the-art Transformer-based models in supporting code completion at different granularity levels, including single tokens, one or multiple entire statements, up to entire code blocks (e.g., the iterated block of a for loop). We experimented with several variants of two recently proposed Transformer-based models, namely RoBERTa and the Text-To-Text Transfer Transformer (T5), for the task of code completion. The achieved results show that Transformer-based models, and in particular the T5, represent a viable solution for code completion, with perfect predictions ranging from ~29%, obtained when asking the model to guess entire blocks, up to ~69%, reached in the simpler scenario of few tokens masked from the same code statement.

cs.SE↗

arXiv2021

A Systematic Literature Review on the Use of Deep Learning in Software Engineering Research

An increasingly popular set of techniques adopted by software engineering (SE) researchers to automate development tasks are those rooted in the concept of Deep Learning (DL). The popularity of such techniques largely stems from their automated feature engineering capabilities, which aid in modeling software artifacts. However, due to the rapid pace at which DL techniques have been adopted, it is difficult to distill the current successes, failures, and opportunities of the current research landscape. In an effort to bring clarity to this crosscutting area of work, from its modern inception to the present, this paper presents a systematic literature review of research at the intersection of SE & DL. The review canvases work appearing in the most prominent SE and DL conferences and journals and spans 128 papers across 23 unique SE tasks. We center our analysis around the components of learning, a set of principles that govern the application of machine learning techniques (ML) to a given problem domain, discussing several aspects of the surveyed work at a granular level. The end result of our analysis is a research roadmap that both delineates the foundations of DL techniques applied to SE research, and highlights likely areas of fertile exploration for the future.

cs.SE↗

arXiv2021

An Empirical Study on the Usage of BERT Models for Code Completion

Code completion is one of the main features of modern Integrated Development Environments (IDEs). Its objective is to speed up code writing by predicting the next code token(s) the developer is likely to write. Research in this area has substantially bolstered the predictive performance of these techniques. However, the support to developers is still limited to the prediction of the next few tokens to type. In this work, we take a step further in this direction by presenting a large-scale empirical study aimed at exploring the capabilities of state-of-the-art deep learning (DL) models in supporting code completion at different granularity levels, including single tokens, one or multiple entire statements, up to entire code blocks (e.g., the iterated block of a for loop). To this aim, we train and test several adapted variants of the recently proposed RoBERTa model, and evaluate its predictions from several perspectives, including: (i) metrics usually adopted when assessing DL generative models (i.e., BLEU score and Levenshtein distance); (ii) the percentage of perfect predictions (i.e., the predicted code snippets that match those written by developers); and (iii) the "semantic" equivalence of the generated code as compared to the one written by developers. The achieved results show that BERT models represent a viable solution for code completion, with perfect predictions ranging from ~7%, obtained when asking the model to guess entire blocks, up to ~58%, reached in the simpler scenario of few tokens masked from the same code statement.

cs.SE↗

arXiv2021

V2S: A Tool for Translating Video Recordings of Mobile App Usages into Replayable Scenarios

Screen recordings are becoming increasingly important as rich software artifacts that inform mobile application development processes. However, the amount of manual effort required to extract information from these graphical artifacts can hinder resource-constrained mobile developers. This paper presents Video2Scenario (V2S), an automated tool that processes video recordings of Android app usages, utilizes neural object detection and image classification techniques to classify the depicted user actions, and translates these actions into a replayable scenario. We conducted a comprehensive evaluation to demonstrate V2S's ability to reproduce recorded scenarios across a range of devices and a diverse set of usage cases and applications. The results indicate that, based on its performance with 175 videos depicting 3,534 GUI-based actions, V2S is accurate in reproducing $\approx$89\% of actions from collected videos.

cs.SE↗

arXiv2021

It Takes Two to Tango: Combining Visual and Textual Information for Detecting Duplicate Video-Based Bug Reports

When a bug manifests in a user-facing application, it is likely to be exposed through the graphical user interface (GUI). Given the importance of visual information to the process of identifying and understanding such bugs, users are increasingly making use of screenshots and screen-recordings as a means to report issues to developers. However, when such information is reported en masse, such as during crowd-sourced testing, managing these artifacts can be a time-consuming process. As the reporting of screen-recordings in particular becomes more popular, developers are likely to face challenges related to manually identifying videos that depict duplicate bugs. Due to their graphical nature, screen-recordings present challenges for automated analysis that preclude the use of current duplicate bug report detection techniques. To overcome these challenges and aid developers in this task, this paper presents Tango, a duplicate detection technique that operates purely on video-based bug reports by leveraging both visual and textual information. Tango combines tailored computer vision techniques, optical character recognition, and text retrieval. We evaluated multiple configurations of Tango in a comprehensive empirical evaluation on 4,860 duplicate detection tasks that involved a total of 180 screen-recordings from six Android apps. Additionally, we conducted a user study investigating the effort required for developers to manually detect duplicate video-based bug reports and compared this to the effort required to use Tango. The results reveal that Tango's optimal configuration is highly effective at detecting duplicate video-based bug reports, accurately ranking target duplicate videos in the top-2 returned results in 83% of the tasks. Additionally, our user study shows that, on average, Tango can reduce developer effort by over 60%, illustrating its practicality.

cs.SE↗

arXiv2021

Studying the Usage of Text-To-Text Transfer Transformer to Support Code-Related Tasks

Deep learning (DL) techniques are gaining more and more attention in the software engineering community. They have been used to support several code-related tasks, such as automatic bug fixing and code comments generation. Recent studies in the Natural Language Processing (NLP) field have shown that the Text-To-Text Transfer Transformer (T5) architecture can achieve state-of-the-art performance for a variety of NLP tasks. The basic idea behind T5 is to first pre-train a model on a large and generic dataset using a self-supervised task ( e.g: filling masked words in sentences). Once the model is pre-trained, it is fine-tuned on smaller and specialized datasets, each one related to a specific task ( e.g: language translation, sentence classification). In this paper, we empirically investigate how the T5 model performs when pre-trained and fine-tuned to support code-related tasks. We pre-train a T5 model on a dataset composed of natural language English text and source code. Then, we fine-tune such a model by reusing datasets used in four previous works that used DL techniques to: (i) fix bugs, (ii) inject code mutants, (iii) generate assert statements, and (iv) generate code comments. We compared the performance of this single model with the results reported in the four original papers proposing DL-based solutions for those four tasks. We show that our T5 model, exploiting additional data for the self-supervised pre-training phase, can achieve performance improvements over the four baselines.

cs.SE↗

arXiv2020

Translating Video Recordings of Mobile App Usages into Replayable Scenarios

Screen recordings of mobile applications are easy to obtain and capture a wealth of information pertinent to software developers (e.g., bugs or feature requests), making them a popular mechanism for crowdsourced app feedback. Thus, these videos are becoming a common artifact that developers must manage. In light of unique mobile development constraints, including swift release cycles and rapidly evolving platforms, automated techniques for analyzing all types of rich software artifacts provide benefit to mobile developers. Unfortunately, automatically analyzing screen recordings presents serious challenges, due to their graphical nature, compared to other types of (textual) artifacts. To address these challenges, this paper introduces V2S, a lightweight, automated approach for translating video recordings of Android app usages into replayable scenarios. V2S is based primarily on computer vision techniques and adapts recent solutions for object detection and image classification to detect and classify user actions captured in a video, and convert these into a replayable test scenario. We performed an extensive evaluation of V2S involving 175 videos depicting 3,534 GUI-based actions collected from users exercising features and reproducing bugs from over 80 popular Android apps. Our results illustrate that V2S can accurately replay scenarios from screen recordings, and is capable of reproducing $\approx$ 89% of our collected videos with minimal overhead. A case study with three industrial partners illustrates the potential usefulness of V2S from the viewpoint of developers.

cs.SE↗

arXiv2020

Cold atoms in micromachined waveguides: a new platform for atom-photon interaction

Hybrid quantum devices, incorporating both atoms and photons, can exploit the benefits of both to enable scalable architectures for quantum computing and quantum communication, as well as chip-scale sensors and single-photon sources. Production of such devices depends on the development of an interface between their atomic and photonic components. This should be compact, robust and compatible with existing technologies from both fields. Here we demonstrate such an interface. Cold cesium atoms are trapped inside a transverse, 30 $μ$m diameter through-hole in an optical fiber, created via laser micromachining. When the guided light is on resonance with the cesium $D_2$ line, up to 87% of it is absorbed by the atoms. The corresponding optical depth per unit length is 700 cm$^{-1}$, higher than any reported for a comparable system. This is important for miniaturisation and scalability. The technique can be equally effective in optical waveguide chips and other existing photonic systems, providing a new platform for fundamental research.

physics.atom-ph↗

arXiv2015

Composite pulses for interferometry in a thermal cold atom cloud

Atom interferometric sensors and quantum information processors must maintain coherence while the evolving quantum wavefunction is split, transformed and recombined, but suffer from experimental inhomogeneities and uncertainties in the speeds and paths of these operations. Several error-correction techniques have been proposed to isolate the variable of interest. Here we apply composite pulse methods to velocity-sensitive Raman state manipulation in a freely-expanding thermal atom cloud. We compare several established pulse sequences, and follow the state evolution within them. The agreement between measurements and simple predictions shows the underlying coherence of the atom ensemble, and the inversion infidelity in an 80 micro-Kelvin atom cloud is halved. Composite pulse techniques, especially if tailored for atom interferometric applications, should allow greater interferometer areas, larger atomic samples and longer interaction times, and hence improve the sensitivity of quantum technologies from inertial sensing and clocks to quantum information processors and tests of fundamental physics.

physics.atom-ph↗

arXiv2014

Extended Wigner function formalism for the spatial propagation of particles with internal degrees of freedom

An extended Wigner function formalism is introduced for describing the quantum dynamics of particles with internal degrees of freedom in the presence of spatially inhomogeneous fields. The approach is used for quantitative simulations of molecular beam experiments involving space-spin entanglement, such as the Stern-Gerlach and the Rabi experiment. The formalism allows a graphical visualization of entanglement and decoherence processes.

quant-ph↗

arXiv2013

Trapping of 85Rb atoms by optical pumping between metastable hyperfine states

We describe an atom trapping mechanism based upon differential optical pumping between metastable hyperfine states by partially-displaced laser beams in the absence of a magnetic field. With realistic laser powers, trap spring constants should match or exceed those typical of magneto-optical traps, and highly flexible tailored trap shapes should be achievable. In a proof-of-principle experiment, we have combined a 1D implementation with magneto-optical trapping in the orthogonal directions, capturing ~10000 85Rb atoms. This is an author-created, un-copyedited version of an article published in J. Phys. B: At. Mol. Opt. Phys. IOP Publishing Ltd is not responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at http://iopscience.iop.org/0953-4075/46/21/215003/article.

physics.atom-ph↗

arXiv2013

Stabilized fiber-optic Mach-Zehnder interferometer for carrier-frequency rejection

We have demonstrated stabilization of a fiber-optic Mach-Zehnder interferometer, with a centimeter-scale path difference, to the transmission minimum for the carrier wave of a frequency-modulated laser beam. A time-averaged extinction of 32 dB, limited by the bandwidth of the feedback, was maintained over several hours. The interferometer was used to remove the carrier wave from a 780 nm laser beam that had been phase modulated at 2.7 GHz.

physics.optics↗

arXiv2013

The high-efficiency γ-ray spectroscopy setup γ3 at HIγS

The existing Nuclear Resonance Fluorescence (NRF) setup at the HIγS facility at the Triangle Universities Nuclear Laboratory at Duke University has been extended in order to perform γ-γ coincidence experiments. The new setup combines large volume LaBr3:Ce detectors and high resolution HPGe detectors in a very close geometry to offer high efficiency, high energy resolution as well as high count rate capabilities at the same time. The combination of a highly efficient γ-ray spectroscopy setup with the mono-energetic high-intensity photon beam of HIγS provides a worldwide unique experimental facility to investigate the γ-decay pattern of dipole excitations in atomic nuclei. The performance of the new setup has been assessed by studying the nucleus \sulfur at 8.125 MeV beam energy. The γ-decay branching ratio from the $1^+$ level at 8125.4 keV to the first excited $2^+$ state was determined to 15.7(3)%.

physics.ins-det↗