Compositional data are non-negative data collected in a rectangular matrix with a constant row sum. Due to the non-negativity the focus is on conditional proportions that add up to 1 for each row. A row of conditional proportions is called an observe
d budget. Latent budget analysis (LBA) assumes a mixture of latent budgets that explains the observed budgets. LBA is usually fitted to a contingency table, where the rows are levels of one or more explanatory variables and the columns the levels of a response variable. In prospective studies, there is only knowledge about the explanatory variables of individuals and interest goes out to predicting the response variable. Thus, a form of LBA is needed that has the functionality of prediction. Previous studies proposed a constrained neural network (NN) extension of LBA that was hampered by an unsatisfying prediction ability. Here we propose LBA-NN, a feed forward NN model that yields a similar interpretation to LBA but equips LBA with a better ability of prediction. A stable and plausible interpretation of LBA-NN is obtained through the use of importance plots and table, that show the relative importance of all explanatory variables on the response variable. An LBA-NN-K- means approach that applies K-means clustering on the importance table is used to produce K clusters that are comparable to K latent budgets in LBA. Here we provide different experiments where LBA-NN is implemented and compared with LBA. In our analysis, LBA-NN outperforms LBA in prediction in terms of accuracy, specificity, recall and mean square error. We provide open-source software at GitHub.
Learning from image-text data has demonstrated recent success for many recognition tasks, yet is currently limited to visual features or individual visual concepts such as objects. In this paper, we propose one of the first methods that learn from im
age-sentence pairs to extract a graphical representation of localized objects and their relationships within an image, known as scene graph. To bridge the gap between images and texts, we leverage an off-the-shelf object detector to identify and localize object instances, match labels of detected regions to concepts parsed from captions, and thus create pseudo labels for learning scene graph. Further, we design a Transformer-based model to predict these pseudo labels via a masked token prediction task. Learning from only image-sentence pairs, our model achieves 30% relative gain over a latest method trained with human-annotated unlocalized scene graphs. Our model also shows strong results for weakly and fully supervised scene graph generation. In addition, we explore an open-vocabulary setting for detecting scene graphs, and present the first result for open-set scene graph generation. Our code is available at https://github.com/YiwuZhong/SGG_from_NLS.
Metal vapour vacuum arcs are capable to generate multiply charged metallic ions, which are widely used in fields such as ion deposition, ion thrusters, and ion sources, etc. According to the stationary model of cathode spot, those ions are generated
by electron-impact single ionization in a step-wise manner, which is M -> M+ -> M2+ -> ... mainly. This paper is designed to study quantitatively the role of double ionization M -> M2+ in the breakdown initiation of copper vacuum arcs. A direct simulation Monte Carlo (DSMC) scheme of double ionization is proposed and incorporated into a 2D particle-in-cell (PIC) method. The super-particles of Cu2+ ions generated from different channels are labelled independently in the PIC-DSMC modelling of vacuum arc breakdown. The cathode erosion rate based on PIC modelling is about 40{mu}g/C in arc burning regime, which agrees well with previous experiments. The temporal discharge behaviours such as arc current, arc voltage, and ionization degree of arc plasma, are influenced with or without double ionization negligibly. However, additional Cu2+ ions are generated near the cathode in breakdown initiation from the double ionization channel, with a lower kinetic energy on average. Therefore, the results on spatial distribution and energy spectra of Cu2+ ions are different with or without double ionization. This paper provides a quantitative research method to evaluate the role of multiply ionization in vacuum arcs.
3D teeth reconstruction from X-ray is important for dental diagnosis and many clinical operations. However, no existing work has explored the reconstruction of teeth for a whole cavity from a single panoramic radiograph. Different from single object
reconstruction from photos, this task has the unique challenge of constructing multiple objects at high resolutions. To conquer this task, we develop a novel ConvNet X2Teeth that decomposes the task into teeth localization and single-shape estimation. We also introduce a patch-based training strategy, such that X2Teeth can be end-to-end trained for optimal performance. Extensive experiments show that our method can successfully estimate the 3D structure of the cavity and reflect the details for each tooth. Moreover, X2Teeth achieves a reconstruction IoU of 0.681, which significantly outperforms the encoder-decoder method by $1.71X and the retrieval-based method by $1.52X. Our method can also be promising for other multi-anatomy 3D reconstruction tasks.
In this paper, by performing a two-dimensional particle-in-cell simulation, we investigate magnetic reconnection in the downstream of a quasi-perpendicular shock. The shock is nonstationary, and experiences a cyclic reformation. At the beginning of r
eformation process, the shock front is relatively flat, and part of upstream ions are reflected by the shock front. The reflected ions move upward in the action of Lorentz force, which leads to the upward bending of magnetic field lines at the foot of the shock front, and then a current sheet is formed due to the squeezing of the bending magnetic field lines. The formed current sheet is brought toward the shock front by the solar wind, and the shock front becomes irregular after interacting with the current sheet. Both the current sheet brought by the solar wind and the current sheet associated with the shock front are then fragmented into many small filamentary current sheets. Electron-scale magnetic reconnection may occur in several of these filamentary current sheets when they are convected into the downstream, and magnetic islands are generated. A strong reconnection electric field and energy dissipation are also generated around the X line, and high-speed electron outflow is also formed.
Studies of shocks have long suggested that a shock can undergo cyclically self-reformation in a time scale of ion cyclotron period. This process has been proposed as a primary mechanism for energy dissipation and energetic particle acceleration at sh
ocks. Unambiguous observational evidence, however, has remained elusive. Here, we report direct observations for the self-reformation process of a collisionless, high Mach number, quasi-perpendicular shock using MMS measurements. We find that reflected ions by the old shock ramp form a clear phase-space vortex, which gives rise to a new ramp. The new ramp observed by MMS2 has not yet developed to a mature stage during the self-reformation, and is not strong enough to reflect incident ions. Consequently, these ions are only slightly slowed down and show a flat velocity profile from the new ramp all the way to the old one. The present results provide direct evidence for shock self-reformation, and also shed light on energy dissipation and energetic particle acceleration at collisionless shocks throughout the universe.
Contrastive learning has been widely used to train transformer-based vision-language models for video-text alignment and multi-modal representation learning. This paper presents a new algorithm called Token-Aware Cascade contrastive learning (TACo) t
hat improves contrastive learning using two novel techniques. The first is the token-aware contrastive loss which is computed by taking into account the syntactic classes of words. This is motivated by the observation that for a video-text pair, the content words in the text, such as nouns and verbs, are more likely to be aligned with the visual contents in the video than the function words. Second, a cascade sampling method is applied to generate a small set of hard negative examples for efficient loss estimation for multi-modal fusion layers. To validate the effectiveness of TACo, in our experiments we finetune pretrained models for a set of downstream tasks including text-video retrieval (YouCook2, MSR-VTT and ActivityNet), video action step localization (CrossTask), video action segmentation (COIN). The results show that our models attain consistent improvements across different experimental settings over previous methods, setting new state-of-the-art on three public text-video retrieval benchmarks of YouCook2, MSR-VTT and ActivityNet.
One-shot neural architecture search (NAS) applies weight-sharing supernet to reduce the unaffordable computation overhead of automated architecture designing. However, the weight-sharing technique worsens the ranking consistency of performance due to
the interferences between different candidate networks. To address this issue, we propose a candidates enhancement method and progressive training pipeline to improve the ranking correlation of supernet. Specifically, we carefully redesign the sub-networks in the supernet and map the original supernet to a new one of high capacity. In addition, we gradually add narrow branches of supernet to reduce the degree of weight sharing which effectively alleviates the mutual interference between sub-networks. Finally, our method ranks the 1st place in the Supernet Track of CVPR2021 1st Lightweight NAS Challenge.
Accuracy predictor is trained to predict the validation accuracy of an network from its architecture encoding. It can effectively assist in designing networks and improving Neural Architecture Search(NAS) efficiency. However, a high-performance predi
ctor depends on adequate trainning samples, which requires unaffordable computation overhead. To alleviate this problem, we propose a novel framework to train an accuracy predictor under few training samples. The framework consists ofdata augmentation methods and an ensemble learning algorithm. The data augmentation methods calibrate weak labels and inject noise to feature space. The ensemble learning algorithm, termed cascade bagging, trains two-level models by sampling data and features. In the end, the advantages of above methods are proved in the Performance Prediciton Track of CVPR2021 1st Lightweight NAS Challenge. Our code is made public at: https://github.com/dlongry/Solutionto-CVPR2021-NAS-Track2.
The benefits that arise from the adoption of a systems engineering approach to the design of engineered systems are well understood and documented. However , with software systems, different approaches are required given the changeability of requirem
ents and the malleability of software. With the design of industrial cyber-physical systems, one is confronted with the challenge of designing engineered systems that have a significant software component. Furthermore, that software component must be able to seamlessly interact with both the enterprises business systems and industrial systems. In this paper, we present Janus, which together with the GORITE BDI agent framework, provides a methodology for the design of agent-based industrial cyber-physical systems. Central to the Janus approach is the development of a logical architecture as in traditional systems engineering and then the allocation of the logical requirements to a BDI (Belief Desire Intention) agent architecture which is derived from the physical architecture for the system. Janus has its origins in product manufacturing; in this paper, we apply it to the problem of Fault Location, Isolation and Service Restoration (FLISR) for power substations.
