Search

This paper provides a comprehensive analysis of Generative AI (GenAI) potential to revolutionise software coding through increased efficiency and reduced time span for writing code. It acknowledges the transformative power of GenAI in software code generation, while also cautioning against the inherent risks of bias and errors if left unchecked. Emphasising the irreplaceable value of traditional programming, it posits that GenAI is not a replacement but a complementary tool for writing software code. Ethical considerations are paramount with the paper advocating for stringent ethical guidelines to ensure GenAI serves the greater good and does not compromise on accountability in writing software code. It suggests a balanced approach, combining human oversight with AI's capabilities, to mitigate risks and enhance reliability. The paper concludes by proposing guidelines for GenAI utilisation in coding, which will empower developers to navigate its complexities and employ it responsibly. This approach addresses current ethical concerns and sets a foundation for the judicious use of GenAI in the future, ensuring its benefits are harnessed effectively while maintaining moral integrity.

Automatic Speech Recognition (ASR) is an active field of research due to its large number of applications and the proliferation of interfaces or computing devices that can support speech processing. However, the bulk of applications are based on well-resourced languages that overshadow under-resourced ones. Yet, ASR represents an undeniable means to promote such languages, especially when designing human-to-human or human-to-machine systems involving illiterate people. An approach to design an ASR system targeting under-resourced languages is to start with a limited vocabulary. ASR using a limited vocabulary is a subset of the speech recognition problem that focuses on the recognition of a small number of words or sentences. This paper aims to provide a comprehensive view of mechanisms behind ASR systems as well as techniques, tools, projects, recent contributions, and possible future directions in ASR using a limited vocabulary. This work consequently provides a way forward when designing an ASR system using limited vocabulary. Although an emphasis is put on limited vocabulary, most of the tools and techniques reported in this survey can be applied to ASR systems in general., Comment: 24 pages, 9 figures, 6 tables, submitted to Elsevier for possible publication

Extracting a connectome from an electron microscopy (EM) data set requires identification of neurons and determination of synapses between neurons. As manual extraction of this information is very time-consuming, there has been extensive research effort to automatically segment the neurons to help guide and eventually replace manual tracing. Until recently, there has been comparatively less research on automatically detecting the actual synapses between neurons. This discrepancy can, in part, be attributed to several factors: obtaining neuronal shapes is a prerequisite first step in extracting a connectome, manual tracing is much more time-consuming than annotating synapses, and neuronal contact area can be used as a proxy for synapses in determining connections. However, recent research has demonstrated that contact area alone is not a sufficient predictor of synaptic connection. Moreover, as segmentation has improved, we have observed that synapse annotation is consuming a more significant fraction of overall reconstruction time. This ratio will only get worse as segmentation improves, gating overall possible speed-up. Therefore, we address this problem by developing algorithms that automatically detect pre-synaptic neurons and their post-synaptic partners. In particular, pre-synaptic structures are detected using a Deep and Wide Multiscale Recursive Network, and post-synaptic partners are detected using a MLP with features conditioned on the local segmentation. This work is novel because it requires minimal amount of training, leverages advances in image segmentation directly, and provides a complete solution for polyadic synapse detection. We further introduce novel metrics to evaluate our algorithm on connectomes of meaningful size. These metrics demonstrate that complete automatic prediction can be used to effectively characterize most connectivity correctly.

The promise of extracting connectomes and performing useful analysis on large electron microscopy (EM) datasets has been an elusive dream for many years. Tracing in even the smallest portions of neuropil requires copious human annotation, the rate-limiting step for generating a connectome. While a combination of improved imaging and automatic segmentation will lead to the analysis of increasingly large volumes, machines still fail to reach the quality of human tracers. Unfortunately, small errors in image segmentation can lead to catastrophic distortions of the connectome. In this paper, to analyze very large datasets, we explore different mechanisms that are less sensitive to errors in automation. Namely, we advocate and deploy extensive synapse detection on the entire antennal lobe (AL) neuropil in the brain of the fruit fly Drosophila, a region much larger than any densely annotated to date. The resulting synapse point cloud produced is invaluable for determining compartment boundaries in the AL and choosing specific regions for subsequent analysis. We introduce our methodology in this paper for region selection and show both manual and automatic synapse annotation results. Finally, we note the correspondence between image datasets obtained using the synaptic marker, antibody nc82, and our datasets enabling registration between light and EM image modalities.

A spacecraft pushed by radiation has the major advantage that the power source is not included in the accelerated mass, making it the preferred technique for reaching relativistic speeds. There are two main technical challenges. First, to get significant acceleration, the sail must be both extremely light weight and capable of operating at high intensities of the incident beam and the resulting high temperatures. Second, the transmitter must emit high power beams through huge apertures, many kilometers in diameter, in order to focus radiation on the sail across the long distances needed to achieve high final speeds. Existing proposals for the sail use carbon or aluminum films, but aluminum is limited by a low melting point, and both have low mechanical strength requiring either a distributed payload or complex rigging. We propose here a graphene sail, which offers high absorption per unit weight, high temperature operation, and the mechanical strength to support simple rigging to a lumped mass payload. For the transmitter, existing proposals use a compact high power source, and focus the energy with a large (hundreds to thousands of km) space-based lens. Existing optical drive proposals also require launch from the outer solar system, have severe pointing restrictions, and require difficult maneuvering of the beam source. Instead we propose an active Fresnel lens, allowing smaller apertures of less mass, easier pointing with fewer restrictions, and probe launch from the inner solar system. The technologies for both the sail and the transmitter are already under development for other reasons. Worked examples, physically smaller and less massive than those suggested so far, range from a 1kg payload launched to 10\% of the speed of light by a transmitter only 25 times the mass of ISS, to a larger system that can launch a 1000 kg payload to 50\% of the speed of light., Comment: 14 pages, 6 figures, 3 tables. V2 includes additional references

The most established method of reconstructing neural circuits from animals involves slicing tissue very thin, then taking mosaics of electron microscope (EM) images. To trace neurons across different images and through different sections, these images must be accurately aligned, both with the others in the same section and to the sections above and below. Unfortunately, sectioning and imaging are not ideal processes - some of the problems that make alignment difficult include lens distortion, tissue shrinkage during imaging, tears and folds in the sectioned tissue, and dust and other artifacts. In addition the data sets are large (hundreds of thousands of images) and each image must be aligned with many neighbors, so the process must be automated and reliable. This paper discusses methods of dealing with these problems, with numeric results describing the accuracy of the resulting alignments., Comment: 23 pages, 10 figures

Interstellar radar is a potential intermediate step between passive observation of exoplanets and interstellar exploratory missions. Compared to passive observation, it has the traditional advantages of radar astronomy. It can measure surface characteristics, determine spin rates and axes, provide extremely accurate ranges, construct maps of planets, distinguish liquid from solid surfaces, find rings and moons, and penetrate clouds. It can do this even for planets close to the parent star. Compared to interstellar travel or probes, it also offers significant advantages. The technology required to build such a radar already exists, radar can return results within a human lifetime, and a single facility can investigate thousands of planetary systems. The cost, although high, is within the reach of Earth's economy, so it is cheaper as well., Comment: 6 pages. no figures; Second version - found previous work and included it; report range for Earth size planets; change proposed polarization to circular

Because of lunar librations, the retroreflectors left on the moon do not, in general, face directly at the Earth. Usually this is regarded as a disadvantage. It results in a spread of arrival times, because each cube that comprises the retroreflector is at a slightly different distance from the earth. However, we can turn this same effect into an advantage. Using pulses and detectors somewhat faster than those currently used for lunar ranging, we can resolve at least some of the structure of a retroreflector, at least when the libration angles are large. This additional structure in the transfer function means that a unique mm level fit can be obtained with many fewer photons. Fitting to the expected reflector transfer function in general requires fewer photons than straight averaging, and smoothly reduces to averaging in the cases where no structure can be resolved. The gains from resolving the reflectors are largest at large libration angles, exactly the case where averaging is most inefficient. In these cases the number of photons needed can be reduced by an order of magnitude or more. Analysis shows that angles for which the gain is very high happen several times each month, with the details depending on the exact librations. Experimental validation of this technique should be possible with existing SLR stations and mockups of the lunar reflectors., Comment: 12 pages; 11 eps figures

In the situation of two electro-statically coupled conductors a current in one conductor may induce a current in the other one. We will study this phenomenon, called Coulomb drag, in the Hubbard chain where the two ``conductors'' are given by fermions with different spin orientation. With the aid of a Monte Carlo (MC) approach which we presented in a recent paper we calculate the Transconductance in different variants of the Hubbard chain (with/without impurity and additional [long-ranged] interactions) for different fillings., Comment: 9 pages, 11 figures

Anderson, et al., find the measured trajectories of Pioneer 10 and 11 spacecraft deviate from the trajectories computed from known forces acting on them. This unmodelled acceleration can be accounted for by non-isotropic radiation of spacecraft heat. Various forms of non-isotropic radiation were proposed by Katz, Murphy, and Scheffer, but Anderson, et al. felt that none of these could explain the observed effect. This paper calculates the known effects in more detail and considers new sources of radiation, all based on spacecraft construction. These effects are then modelled over the duration of the experiment. The model provides a reasonable fit to the acceleration from its appearance at a heliocentric distance of 5 AU to the last measurement at 71 AU, but overpredicts by 9% the decrease in acceleration between intervals I and III of the Pioneer 10 observations. (For comparison, the two different measurements of the effect (SIGMA and CHASMP) themselves differ by 4% in interval III.) In any case, by accounting for the bulk of the acceleration, the proposed mechanism makes it much more likely that the entire effect can be explained without the need for new physics., Comment: LaTex; 5 pages, 2 figures, to be submitted to PRL

Anderson, et al. find the measured trajectories of Pioneer 10 and 11 spacecraft deviate from the trajectories computed from known forces acting on them. This unmodelled acceleration (and the less well known, but similar, unmodelled torque) can be accounted for by non-isotropic radiation of spacecraft heat. Various forms of non-isotropic radiation were proposed by Katz, Murphy, and Scheffer, but Anderson, et al. felt that none of these could explain the observed effect. This paper calculates the known effects in more detail and considers new sources of radiation, all based on spacecraft construction. These effects are then modelled over the duration of the experiment. The model reproduces the acceleration from its appearance at a heliocentric distance of 5 AU to the last measurement at 71 AU to within 10 percent. However, it predicts a larger decrease in acceleration between intervals I and III of the Pioneer 10 observations than is observed. This is a 2 sigma discrepancy from the average of the three analyses (SIGMA, CHASMP, and Markwardt). A more complex (but more speculative) model provides a somewhat better fit. Radiation forces can also plausibly explain the previously unmodelled torques, including the spindown of Pioneer 10 that is directly proportional to spacecraft bus heat, and the slow but constant spin-up of Pioneer 11. In any case, by accounting for the bulk of the acceleration, the proposed mechanism makes it much more likely that the entire effect can be explained without the need for new physics., Comment: Final minor changes for publication - added explanation of acronyms, added to RTG asymmetry argument.. Was: 12 pages, 9 figures, major revision. Added discussion of gas leaks and spin history, a radiation based explanation of spin changes, and references to re-analysis of Markwardt. Fixed radio forces, tuned models. Was: 7 pages, 5 figures; added liklihood calculations in body and abstract per suggestion

Anderson, et al., find the measured trajectories of Pioneer 10 and 11 spacecraft deviate from the trajectories computed from known forces acting on them. This unmodelled acceleration can be accounted for by non-isotropic radiation of spacecraft heat. This explanation was first proposed by Murphy, but Anderson, et al. felt it could not explain the observed effect. This paper includes new calculations on the expected magnitude of this effect, based on the relative emissivities of the different sides of the spacecraft, as estimated from the known spacecraft construction. The calculations indicate the proposed effect can account for most, if not all, of the unmodelled acceleration., Comment: 3 pages, revtex4

We observe the eruption of a large-scale (approx.300,000 km) quiet-region solar filament, leading to an Earth-directed "halo" coronal mass ejection (CME). We use coronal imaging data in EUV from the EUV Imaging Telescope (EIT) on the Solar and Heliospheric Observatory (SOHO) satellite, and in soft X-rays (SXRs) from the Soft X-ray Telescope (SXT) on the Yohkoh satellite. We also use spectroscopic data from the Coronal Diagnostic Spectrometer (CDS), magnetic data from the Michelson Doppler Imager (MDI), and white-light coronal data from the Large Angle and Spectrometric Coronagraph Experiment (LASCO), all on SOHO. Initially the filament shows a slow (approx.1 km/s projected against the solar disk) and approximately constant-velocity rise for about 6 hours, before erupting rapidly, reaching a velocity of approx. 8 km/s over the next approx. 25 min. CDS Doppler data show Earth-directed filament velocities ranging from < 20 km/s (the noise limit) during the slow-rise phase, to approx. 100 km/s-1 early in the eruption. Beginning within 10 hours prior to the start of the slow rise, localized new magnetic flux emerged near one end of the filament. Near the start of and during the slow-rise phase, SXR microflaring occurred repeatedly at the flux-emergence site, in conjunction with the development of a fan of SXR illumination of the magnetic arcade over the filament. The SXR microflares, development of the SXR fan, and motion of the slow-rising filament are all consistent with "tether-weakening" reconnection occurring between the newly-emerging flux and the overlying arcade field containing the filament field. The microflares and fan structure are not prominent in EUV, and would not have been detected without the SXR data. Standard "twin dimmings" occur near the location of the filament, and "remote dimmings" and "brightenings" occur further removed from the filament.

The NASA White Sands Test Facility (WSTF) routinely operates hypergolic propulsion systems. Some of the onsite activities include performing long duration studies on the operational life of these systems. A few of them have been in use for over twenty years. During this span of time contamination has built up in the propellant and some of the distribution infrastructure. This study investigated the nature of this contamination, the pathology of its generation, and developed a process for removal of the contamination that was cost efficient with minimal waste generation.