Search For Objects Research Articles

Do Helpful Mothers Help? Effects of Maternal Scaffolding and Infant Engagement on Cognitive Performance.

Infants are highly social and much early learning takes place in a social context during interactions with caregivers. Previous research shows that social scaffolding – responsive parenting and joint attention – can confer benefits for infants’ long-term development and learning. However, little previous research has examined whether dynamic (moment-to-moment) adaptations in adults’ social scaffolding are able to produce immediate effects on infants’ performance. Here we ask whether infants’ success on an object search task is more strongly influenced by maternal behavior, including dynamic changes in response behavior, or by fluctuations in infants’ own engagement levels. Thirty-five mother-infant dyads (infants aged 10.8 months, on average) participated in an object search task that was delivered in a naturalistic manner by the child’s mother. Measures of maternal responsiveness (teaching duration; sensitivity) and infant engagement (engagement score; visual attention) were assessed. Mothers varied their task delivery trial by trial, but neither measure of maternal responsiveness significantly predicted infants’ success in performing the search task. Rather, infants’ own level of engagement was the sole significant predictor of accuracy. These results indicate that while parental scaffolding is offered spontaneously (and is undoubtedly crucial for development), in this context children’s endogenous engagement proved to be a more powerful determinant of task success. Future work should explore this interplay between parental and child-internal factors in other learning and social contexts.

Weak-U(1) × strong-U(1) effective gauge field theories and electron-monopole scattering

We present a gauge and Lorentz invariant model for the scattering of matter off magnetic poles, which justifies the presence of velocity-dependent magnetic charges as an effective description of either the behaviour of monopoles in scattering with matter or their production from matter particles at colliders. Hence, in such an approach, perturbativity of the magnetic charge is ensured for relative low velocities of monopoles with respect to matter particles. The model employs a ${\rm U(1)}_{\rm weak} \times {\rm U(1)}_{\rm strong}$ effective gauge field theory under which electrons and monopoles (assumed to be fermions) are appropriately charged. The non-perturbative quantum effects of the strongly coupled sector of the theory lead to dressed effective couplings of the monopole/dyon with the electromagnetic photon, due to non-trivial wave-function renormalization effects. For slowly-moving monopole/dyons, such effects lead to weak coupling, thus turning the bare non-perturbative magnetic charge, which is large due to the Dirac/Schwinger quantization rule, into a perturbative effective, velocity-("$\beta$")-dependent magnetic coupling. Our work thus offers formal support to previous conjectural studies, employing effective U(1)-electromagnetic gauge field theories for the description of monopole production from Standard Model matter, which are used in contemporary collider searches of such objects. This work necessarily pertains to composite monopoles, as seems to be the case of all known monopoles so far, that are solutions of specific particle physics models. This is a consequence of the fact that the wave-function renormalization of the (slowly-moving) monopole fermion turns out to violate unitarity bounds that would characterise asymptotic elementary particle states.

ПРИМЕНЕНИЕ ОБЪЕКТНОЙ МОДЕЛИ ДЛЯ КОНСТРУИРОВАНИЯ ТЯГОВОГО ПРИВОДА ЛОКОМОТИВА

The paper reports the consideration of the problem in choice and use of the locomotive traction drive design procedure giving possibilities both in properties and parts modeling, and in the automated identification of design solutions of latter. The object model of a design solution as a structure including the totality of object descriptions allowing the synthesis of engineering solution models as a multitude of interrelated elements described in the libraries of well-known designs. 
 The method differs from the earlier ones in which the creation of new solutions is stipulated by the definition of a prototype and then by its development in accordance with the tendencies revealed empirically. In the method developed the synthesis of a design object model is based on engineering system classifications. 
 The integrated object model of a traction drive involves two parts: first, the multitude hierarchy of the traction drive description for different methods of an engineering solution schematization, secondly, a library which has descriptions of standard objects in the form of a hierarchal structure of functional interaction between elements (I-graph). The search of similar objects in the library takes place at the level of the definition of drive functional elements with the aid of the matrices of similarity measures. There are given offers for the increase of reliability and loading capacity of a rubber-cord clutch of a locomotive traction drive. There are obtained three patents for utility models and an invention patent on the solutions offered.

Characteristic Sounds Facilitate Object Search in Real-Life Scenes.

Real-world events do not only provide temporally and spatially correlated information across the senses, but also semantic correspondences about object identity. Prior research has shown that object sounds can enhance detection, identification, and search performance of semantically consistent visual targets. However, these effects are always demonstrated in simple and stereotyped displays that lack ecological validity. In order to address identity-based cross-modal relationships in real-world scenarios, we designed a visual search task using complex, dynamic scenes. Participants searched for objects in video clips recorded from real-life scenes. Auditory cues, embedded in the background sounds, could be target-consistent, distracter-consistent, neutral, or just absent. We found that, in these naturalistic scenes, characteristic sounds improve visual search for task-relevant objects but fail to increase the salience of irrelevant distracters. Our findings generalize previous results on object-based cross-modal interactions with simple stimuli and shed light upon how audio–visual semantically congruent relationships play out in real-life contexts.

Fast Similarity Search for Graphs by Edit Distance

This survey article considers index structures for fast similarity search for objects represented by trees and graphs. Edit distance is used as a measure of similarity. The execution of exact similarity search queries is considered. Algorithms based on the filter-and-refine strategy using inverted indexing are mainly presented. Algorithms for exact calculation of the graph edit distance and its lower and upper bounds are also considered.

Search for compact dark matter objects in the solar system with LIGO data

Dark matter could be composed of compact dark objects (CDOs). A close binary of CDOs orbiting in the interior of solar system bodies can be a loud source of gravitational waves (GWs) for the LIGO and VIRGO detectors. We perform the first search ever for this type of signal and rule out close binaries, with separations of order 300 m, orbiting near the center of the Sun with GW frequencies (twice the orbital frequency) between 50 and 550 Hz and CDO masses above ≈10−9M⊙. This mass limit is eight orders of magnitude lower than the mass probed in a LIGO search at extra galactic distances.

ТЕОРЕТИЧНІ ПІДХОДИ ДО ВИЗНАЧЕННЯ ПОНЯТЬ «ІНФОРМАЦІЙНИЙ ПРОСТІР», «ІНФОРМАЦІЙНЕ СЕРЕДОВИЩЕ», «ІНФОРМАЦІЙНО-КОМУНІКАТИВНЕ СЕРЕДОВИЩЕ

Importance. The rapid development of the modern information society leads to the emergence of new phenomena of objective reality, which requires their clear definition, description, as well as constant refinement and additions. The terms of the information industry become widely interdisciplinary and continue to be studied actively. The result of this is the diversity and opposition to the views on certain concepts, ambiguity in the interpretation, excessive generalization, which leads to their complete or partial mixing, identification, violation in the relationship between them. Objective. The purpose of the article is to differentiate the notion of «information space», «information environment», «informational and communicative environment», to determine their essence, to find out how they co-exist in the system of managerial activity.Results. After analyzing the content of the notions of «information space» and «information environment», we found that the main difference between them is that the information space represents a set of all available to society, the state, the Union of states of information resources, knowledge, information networks, systems, technologies, processes related to the search, processing and storage of information, objects and subjects of information activities, as well as the legal and regulatory framework that controls the use of certain information technology but the information environment is a set of certain information technologies that are part of the information space and are used by a specific unit in the management system. The notion of «informational and communicative environment» stems from the «information environment», its main feature is the provision of a communicative function in the process of information interaction between actors of information activity.

MoodleREC: A recommendation system for creating courses using the moodle e-learning platform

The field of education has never been indifferent to the new technologies, and eventually to the Internet. Technology-Enhanced Learning, progressively, has grown to be the area for research and practice on the application of information and communication technologies to teaching and learning. In particular for the teaching activity, the numerous standard compliant Learning Object Repositories available via the Internet, and Open Educational Resources repositories, provide formidable support to teachers when they need to develop a course that can also make use of already available learning materials. The search and selection of Learning Objects, however, can be an inherently complex operation involving accessing various repositories, each potentially involving different software tools, and different organization and specification formats for the learning resources. This complexity may hinder the very success of an e-learning course. Cross-repository aggregators, i.e., systems that can roam through different repositories to satisfy the user's/teacher's query, can help to reduce such complexity, although problems of course delivery may remain. This paper proposes a hybrid recommender system, MoodleRec, implemented as a plug-in of the Moodle Learning Management System. MoodleRec can sort through a set of supported standard compliant Learning Object Repositories, and suggest a ranked list of Learning Objects following a simple keyword-based query. The various recommendation strategies operate on two levels. First, a ranked list of Learning Objects is created, ordered by their correspondence to the query, and by their quality, as indicated by the repository of origin. Social generated features are then used to show the teacher how the Learning Objects listed have been exploited in other courses. A real life experimental study is also presented, and the validity of the MoodleRec approach discussed.

Search for Subsolar Mass Ultracompact Binaries in Advanced LIGO’s Second Observing Run

We present a search for subsolar mass ultracompact objects in data obtained during Advanced LIGO's second observing run. In contrast to a previous search of Advanced LIGO data from the first observing run, this search includes the effects of component spin on the gravitational waveform. We identify no viable gravitational-wave candidates consistent with subsolar mass ultracompact binaries with at least one component between 0.2 M_{⊙}-1.0 M_{⊙}. We use the null result to constrain the binary merger rate of (0.2 M_{⊙}, 0.2 M_{⊙}) binaries to be less than 3.7×10^{5} Gpc^{-3} yr^{-1} and the binary merger rate of (1.0 M_{⊙}, 1.0 M_{⊙}) binaries to be less than 5.2×10^{3} Gpc^{-3} yr^{-1}. Subsolar mass ultracompact objects are not expected to form via known stellar evolution channels, though it has been suggested that primordial density fluctuations or particle dark matter with cooling mechanisms and/or nuclear interactions could form black holes with subsolar masses. Assuming a particular primordial black hole (PBH) formation model, we constrain a population of merging 0.2 M_{⊙} black holes to account for less than 16% of the dark matter density and a population of merging 1.0 M_{⊙} black holes to account for less than 2% of the dark matter density. We discuss how constraints on the merger rate and dark matter fraction may be extended to arbitrary black hole population models that predict subsolar mass binaries.

The integration of visual and target signals in V4 and IT during visual object search

Searching for a specific visual object requires our brain to compare the items in view with a remembered representation of the sought target to determine whether a target match is present. This comparison is thought to be implemented, in part, via the combination of top-down modulations reflecting target identity with feed-forward visual representations. However, it remains unclear whether top-down signals are integrated at a single locus within the ventral visual pathway (e.g., V4) or at multiple stages [e.g., both V4 and inferotemporal cortex (IT)]. To investigate, we recorded neural responses in V4 and IT as rhesus monkeys performed a task that required them to identify when a target object appeared across variation in position, size, and background context. We found nonvisual, task-specific signals in both V4 and IT. To evaluate whether V4 was the only locus for the integration of top-down signals, we evaluated several feed-forward accounts of processing from V4 to IT, including a model in which IT preferentially sampled from the best V4 units and a model that allowed for nonlinear IT computation. IT task-specific modulation was not accounted for by any of these feed-forward descriptions, suggesting that during object search, top-down signals are integrated directly within IT.NEW & NOTEWORTHY To find specific objects, the brain must integrate top-down, target-specific signals with visual information about objects in view. However, the exact route of this integration in the ventral visual pathway is unclear. In the first study to systematically compare V4 and inferotemporal cortex (IT) during an invariant object search task, we demonstrate that top-down signals found in IT cannot be described as being inherited from V4 but rather must be integrated directly within IT itself.

Принятие решений на основе вывода по прецедентам в моделировании месторождений нефти и газа

This article describes a case-based reasoning (CBR) method as a decision-making tool in modeling oil and gas fields, its existing and potential application in the industry. The main direction for the application of the CBR method is the search for analogous objects for the design of oil and gas field development. The current engineering practice involves a vaguely formalized method of analogies, which does not allow defining the object as much as possible authentically the analogue that does not cause errors. The analogue object serves not only as a source of ready-made near optimal design solutions, but also as additional information about the object of development and key decisions in modeling hydrocarbon fields.<br> This paper considers the CBR method as the main tool for finding analogue objects, the main methods of extracting precedents from the database, and gives an idea of the object of development as a precedent. Proceeding from the peculiarities of presenting the object of development as a precedent and the peculiarities of applying the methods of extracting precedents, the authors have developed the concept of searching for analogue objects. In its implementation, it will allow for a different degree of information content of precedents stored in the database and will accelerate the procedure of extracting precedents from the database. The principal novelty is that the presented conceptual scheme allows using the methods of extracting precedents in the conditions of insufficient input data, which is important for the design of oil and gas fields.

Additional Value of Using Satellite-Based Soil Moisture and Two Sources of Groundwater Data for Hydrological Model Calibration

Although the complexity of physically-based models continues to increase, they still need to be calibrated. In recent years, there has been an increasing interest in using new satellite technologies and products with high resolution in model evaluations and decision-making. The aim of this study is to investigate the value of different remote sensing products and groundwater level measurements in the temporal calibration of a well-known hydrologic model i.e., Hydrologiska Bryåns Vattenbalansavdelning (HBV). This has rarely been done for conceptual models, as satellite data are often used in the spatial calibration of the distributed models. Three different soil moisture products from the European Space Agency Climate Change Initiative Soil Measure (ESA CCI SM v04.4), The Advanced Microwave Scanning Radiometer on the Earth Observing System (EOS) Aqua satellite (AMSR-E), soil moisture active passive (SMAP), and total water storage anomalies from Gravity Recovery and Climate Experiment (GRACE) are collected and spatially averaged over the Moselle River Basin in Germany and France. Different combinations of objective functions and search algorithms, all targeting a good fit between observed and simulated streamflow, groundwater and soil moisture, are used to analyze the contribution of each individual source of information. Firstly, the most important parameters are selected using sensitivity analysis, and then these parameters are included in a subsequent model calibration. The results of our multi-objective calibration reveal a substantial contribution of remote sensing products to the lumped model calibration, even if their spatially-distributed information is lost during the spatial aggregation. Inclusion of new observations, such as groundwater levels from wells and remotely sensed soil moisture to the calibration improves the model’s physical behavior, while it keeps a reasonable water balance that is the key objective of every hydrologic model.

Efficient Dynamic Object Search in Home Environment by Mobile Robot: A Priori Knowledge-Based Approach

The capability of object search is a prerequisite for mobile robot to perform everyday tasks in the home environment. Due to the dynamic nature of object and the particularity of home environment, it still remains a challenging problem for mobile robot to find the dynamic target object with the low total cost. To address this problem, a priori knowledge-based approach for dynamic object search is presented in this paper. Inspired by human search process, the common sense knowledge about typical spatial location and the relationship between objects is modeled as a priori knowledge. On this basis, we propose a novel search strategy to improve the search efficiency of mobile robot at the scale of the entire home environment. Unlike the existing methods, a trade-off between the inferred spatial location knowledge and the distance of robot position and candidate room is considered, which permits robot to prioritize the search effort for the candidate room in which the target object is most likely to be found. Also, the search space inside the room is further reduced by using a priori knowledge-guided search heuristics. Meanwhile, the update of knowledge is also investigated to maintain its reliability. The proposed approach is implemented in both simulation and real environments, and evaluated through extensive experiments. The experimental results are provided to demonstrate the feasibility and efficiency of our proposal.

VEXAS: VISTA EXtension to Auxiliary Surveys

Context.We present the first public data release of the VISTA EXtension to Auxiliary Surveys (VEXAS), comprising nine cross-matched multi-wavelength photometric catalogues where each object has a match in at least two surveys.Aims.Our aim is to provide spatial coverage that is as uniform as possible in the multi-wavelength sky and to provide the astronomical community with reference magnitudes and colours for various scientific uses: object classification (e.g. quasars, galaxies, and stars; high-zgalaxies, white dwarfs); photometric redshifts of large galaxy samples; searches of exotic objects (e.g. extremely red objects and lensed quasars).Methods.We cross-matched the wide-field VISTA catalogues (the VISTA Hemisphere Survey and the VISTA Kilo Degree Infrared Galaxy Survey) with the AllWISE mid-infrared Survey, requiring a match within 10″. We have further matched this table with X-ray and radio data (ROSAT, XMM, SUMSS). We also performed a second cross-match between VISTA and AllWISE, with a smaller matching radius (3″), including WISE magnitudes. We then cross-matched this resulting table (≈138 × 106objects) with three photometric wide-sky optical deep surveys (DES, SkyMapper, PanSTARRS). We finally included matches to objects with spectroscopic follow-up by the SDSS and 6dFGS.Results.To demonstrate the power of all-sky multi-wavelength cross-match tables, we show two examples of scientific applications of VEXAS, in particular using the publicly released tables to discover strong gravitational lenses (beyond the reach of previous searches) and to build a statistically large sample of extremely red objects.Conclusions.The VEXAS catalogue is currently the widest and deepest public optical-to-IR photometric and spectroscopic database in the southern hemisphere.

Future considerations in prosthetic urology.

Since their popularization, genitourinary prosthetics have remained a gold-standard therapy for the treatment of erectile dysfunction and stress urinary incontinence and in cases of testicular loss or dysfunction. They have also represented an area of significant innovation, which has contributed to excellent long-term outcomes. Given this history, the objective of the current review was to provide a 5–10-year outlook on anticipated trends and developments in the field of genitourinary prosthetics. To accomplish this objective, a PubMed and patent search was performed of topics relating to penile and testicular prostheses and urinary sphincters. In regard to penile prostheses, findings demonstrated several new concepts including temperature-sensitive alloys, automated pumps, devices designed specifically for neophalluses, and improved malleable designs. With artificial urinary sphincters, new concepts include the ability to add or remove fluid from an existing system, two-piece systems, and new mechanisms to occlude the urethra. For testicular prosthetics, future implementations may not only better replicate the feel of a biological testicle but also add endocrinological functions. Beyond device innovation, the future of prosthetics is also one of expanding geographic boundaries, which necessitates variable cost modeling and regulatory considerations. Surgical trends are also changing, with a greater emphasis on nonnarcotic, postoperative pain control, outpatient surgeries, and adjunctive techniques to lengthen the penis and address concomitant stress incontinence, among others. Concomitant with device and surgical changes, future considerations also include a greater need for education and training, particularly given the rapid expansion of sexual medicine into developing nations.

The Hubble Catalog of Variables (HCV)

Aims. Over its lifetime and despite not being a survey telescope, the Hubble Space Telescope (HST) has obtained multi-epoch observations by multiple, diverse observing programs, providing the opportunity for a comprehensive variability search aiming to uncover new variables. We have therefore undertaken the task of creating a catalog of variable sources based on archival HST photometry. In particular, we have used version 3 of the Hubble Source Catalog (HSC), which relies on publicly available images obtained with the WFPC2, ACS, and WFC3 instruments onboard the HST. Methods. We adopted magnitude-dependent thresholding in median absolute deviation (a robust measure of light curve scatter) combined with sophisticated preprocessing techniques and visual quality control to identify and validate variable sources observed by Hubble with the same instrument and filter combination five or more times. Results. The Hubble Catalog of Variables (HCV) includes 84 428 candidate variable sources (out of 3.7 million HSC sources that were searched for variability) with V ≤ 27 mag; for 11 115 of them the variability is detected in more than one filter. The data points in the light curves of the variables in the HCV catalog range from five to 120 points (typically having less than ten points); the time baseline ranges from under a day to over 15 years; while ∼8% of all variables have amplitudes in excess of 1 mag. Visual inspection performed on a subset of the candidate variables suggests that at least 80% of the candidate variables that passed our automated quality control are true variable sources rather than spurious detections resulting from blending, residual cosmic rays, and calibration errors. Conclusion. The HCV is the first, homogeneous catalog of variable sources created from the highly diverse, archival HST data and currently is the deepest catalog of variables available. The catalog includes variable stars in our Galaxy and nearby galaxies, as well as transients and variable active galactic nuclei. We expect that the catalog will be a valuable resource for the community. Possible uses include searches for new variable objects of a particular type for population analysis, detection of unique objects worthy of follow-up studies, identification of sources observed at other wavelengths, and photometric characterization of candidate progenitors of supernovae and other transients in nearby galaxies. The catalog is available to the community from the ESA Hubble Science Archive (eHST) at the European Space Astronomy Centre (ESAC) and the Mikulski Archive for Space Telescopes (MAST) at Space Telescope Science Institute (STScI).

Features of investigative actions at the initial stage of the investigation of criminal violations of privacy

The features of the investigative actions carried out at the initial stage of the investigation of criminal privacy violations are considered. It is noted that in the process of selecting investigative actions and determining their sequence, the investigator builds up an investigation algorithm that takes into account the characteristics of the crime and the investigative situation. When choosing the investigative actions at the initial stage of the investigation, it should be borne in mind that they should be orienting and searching. The features of the production of interrogations of the victim and witnesses, a search, seizure and investigative examination of objects and documents are examined. An important source of guidance information is the receipt of information relevant to establishing the circumstances of the crime under investigation from organizations that provide communication services.It is concluded that the high-quality production of investigative actions of the initial stage of the investigation, taking into account the specifics of criminal violations of privacy and the correct assessment of their results, should contribute to the development of versions of the involvement of specific individuals in the commission of the crime.

An adaptive probabilistic atlas for anomalous brain segmentation in MR images.

Automated segmentation of brain structures (objects) in MR three-dimensional (3D) images for quantitative analysis has been a challenge and probabilistic atlases (PAs) are among the most well-succeeded approaches. However, the existing models do not adapt to possible object anomalies due to the presence of a disease or a surgical procedure. Post-processing operation does not solve the problem, for example, tissue classification to detect and remove such anomalies inside the resulting segmentation mask, because segmentation errors on healthy tissues cannot be fixed. Such anomalies very often alter the shape and texture of the brain structures, making them different from the appearance of the model. In this paper, we present an effective and efficient adaptive probabilistic atlas, named AdaPro, to circumvent the problem and evaluate it on a challenging task - the segmentation of the left hemisphere, right hemisphere, and cerebellum, without pons and medulla, in 3D MR-T1 brain images of Epilepsy patients. This task is challenging due to temporal lobe resections, artifacts, and the absence of contrast in some parts between the structures of interest. In AdaPro, we first build one probabilistic atlas per object of interest from a training set with normal 3D images and the corresponding 3D object masks. Second, we incorporate a texture classifier based on convex optimization which dynamically indicates the regions of the target 3D image where the PAs (shape constraints) should be further adapted. This strategy is mathematically more elegant and avoids problems with post-processing. Third, we add a new object-based delineation algorithm based on combinatorial optimization and diffusion filtering. AdaPro can then be used to locate and delineate the objects in the coordinate space of the atlas or of the test image. We also compare AdaPro with three other state-of-the-art methods: an statistical shape model based on synergistic object search and delineation, and two methods based on multi-atlas label fusion (MALF). We evaluate the methods quantitatively on 3D MR-T1 brain images of 2T and 3T from epilepsy patients, before and after temporal lobe resections, and on the template and native coordinate spaces. The results show that AdaPro is considerably faster and consistently more accurate than the baselines with statistical significance in both coordinate spaces. AdaPro can be used as a fast and effective step for brain tissue segmentation and it can also be easily extended to segment subcortical brain structures. By choice of its components, probabilistic atlas, texture classifier, and delineation algorithm, it can also be extended to other organs and imaging modalities.

Intelligent Embedded Vision for Summarization of Multiview Videos in IIoT

Nowadays, video sensors are used on a large scale for various applications, including security monitoring and smart transportation. However, the limited communication bandwidth and storage constraints make it challenging to process such heterogeneous nature of Big Data in real time. Multiview video summarization (MVS) enables us to suppress redundant data in distributed video sensors settings. The existing MVS approaches process video data in offline manner by transmitting them to the local or cloud server for analysis, which requires extra streaming to conduct summarization, huge bandwidth, and are not applicable for integration with industrial Internet of Things (IIoT). This article presents a light-weight convolutional neural network (CNN) and IIoT-based computationally intelligent (CI) MVS framework. Our method uses an IIoT network containing smart devices, Raspberry Pi (RPi) (clients and master) with embedded cameras to capture multiview video data. Each client RPi detects target in frames via light-weight CNN model, analyzes these targets for traffic and crowd density, and searches for suspicious objects to generate alert in the IIoT network. The frames of each client RPi are encoded and transmitted with approximately 17.02% smaller size of each frame to master RPi for final MVS. Empirical analysis shows that our proposed framework can be used in industrial environments for various applications such as security and smart transportation and can be proved beneficial for saving resources. 1 1 [Online]. Available: https://github.com/tanveer-hussain/Embedded-Vision-for-MVS .

Axion boson stars

We study novel solitonic solutions to Einstein-Klein-Gordon theory in the presence of a periodic scalar potential arising in models of axion-like particles. The potential depends on two parameters: the mass of the scalar field ma and the decay constant fa; the standard case of the QCD axion is recovered when ma ∝ 1/fa. When fa → ∞ the solutions reduce to the standard case of “mini” boson stars supported by a massive free scalar field. As the energy scale fa of the scalar self-interactions decreases we unveil several novel features of the solution: new stability branches emerge at high density, giving rise to very compact, radially stable, boson stars. Some of the most compact configurations acquire a photon sphere. When fa is at the GUT scale, a boson star made of QCD axions can have a mass up to ten solar masses and would be more compact than a neutron star. Gravitational-wave searches for these exotic compact objects might provide indirect evidence for ultralight axion-like particles in a region not excluded by the black-hole superradiant instability.