Set Of Prototypes Research Articles

Designing for Embodied Proximity

This paper draws on notions of embodied learning to inform exhibit design that fosters children’s meaningful embodied engagement to successfully unveil science ideas. While children’s interaction in the museum is often hands-on and speaks to the physical emphasis that embodiment brings, observation of children’s spontaneous engagement at a museum’s Water Zone revealed opportunities and barriers to engagement with, and access to, science ideas in terms of what we call ‘embodied proximity’ and ‘embodied dislocation’. Drawing on design considerations from these findings a set of purpose-built prototype exhibits were developed and deployed to examine how they supported children’s embodied exploration of science. The findings highlight key design dimensions that support children’s accessing and making meaning about science through fostering embodied proximity: considering palette of embodied features; applying direct multisensorial experience; developing temporal-positional contiguity; and designing opportunities for communicating experiences through the body.

Feature Consistency-Based Prototype Network for Open-Set Hyperspectral Image Classification.

Hyperspectral image (HSI) classification methods have made great progress in recent years. However, most of these methods are rooted in the closed-set assumption that the class distribution in the training and testing stages is consistent, which cannot handle the unknown class in open-world scenes. In this work, we propose a feature consistency-based prototype network (FCPN) for open-set HSI classification, which is composed of three steps. First, a three-layer convolutional network is designed to extract the discriminative features, where a contrastive clustering module is introduced to enhance the discrimination. Then, the extracted features are used to construct a scalable prototype set. Finally, a prototype-guided open-set module (POSM) is proposed to identify the known samples and unknown samples. Extensive experiments reveal that our method achieves remarkable classification performance over other state-of-the-art classification techniques.

Design of Automatic Seaming Machine for Fire Air Ducts with 3D Model and Its Experiment

The fire air duct is one of the important components of the ventilation system in large buildings, which is is a unit system for ventilation and smoke exhaust. The ventilation system is a customized system with a wide range of applications and a high demand, resulting in the use of multiple sizes and specifications of fire air ducts. At the same time, fire air ducts are generally made of thin galvanized iron sheets, which have poor rigidity and are prone to deformation. For the two above-mentioned reasons, the seaming process of the fire air duct is currently mainly completed by manual tapping, resulting in low processing efficiency, high labor intensity, and high cost. Therefore, it is necessary to design an automated equipment that can complete the fire air duct seaming process. Firstly, a 3D model for automatic seaming machine for fire air ducts to demonstrate the working principle and a double head positioning and clamping mechanism is proposed to improve the pipe rigidity; Secondly, to adapt to most specifications for rectangular fire air ducts, a system was designed that utilizes the two right angle ends of the stitching edge for positioning and clamping without many special toolings. Thirdly, a set of automatic seaming prototype has been manufactured based on the above design ideas. The experimental results on the prototype show that the fire air duct automatic stitching machine can seam a fire air duct within 1 minute. Compared with manual processing, the processing efficiency is improved by more than 80%.

Interpretable3D: An Ad-Hoc Interpretable Classifier for 3D Point Clouds

3D decision-critical tasks urgently require research on explanations to ensure system reliability and transparency. Extensive explanatory research has been conducted on 2D images, but there is a lack in the 3D field. Furthermore, the existing explanations for 3D models are post-hoc and can be misleading, as they separate explanations from the original model. To address these issues, we propose an ad-hoc interpretable classifier for 3D point clouds (i.e., Interpretable3D). As an intuitive case-based classifier, Interpretable3D can provide reliable ad-hoc explanations without any embarrassing nuances. It allows users to understand how queries are embedded within past observations in prototype sets. Interpretable3D has two iterative training steps: 1) updating one prototype with the mean of the embeddings within the same sub-class in Prototype Estimation, and 2) penalizing or rewarding the estimated prototypes in Prototype Optimization. The mean of embeddings has a clear statistical meaning, i.e., class sub-centers. Moreover, we update prototypes with their most similar observations in the last few epochs. Finally, Interpretable3D classifies new samples according to prototypes. We evaluate the performance of Interpretable3D on four popular point cloud models: DGCNN, PointNet2, PointMLP, and PointNeXt. Our Interpretable3D demonstrates comparable or superior performance compared to softmax-based black-box models in the tasks of 3D shape classification and part segmentation. Our code is released at: github.com/FengZicai/Interpretable3D.

Personalized bundle recommendation using preference elicitation and the Choquet integral.

Bundle recommendation aims to generate bundles of associated products that users tend to consume as a whole under certain circumstances. Modeling the bundle utility for users is a non-trivial task, as it requires to account for the potential interdependencies between bundle attributes. To address this challenge, we introduce a new preference-based approach for bundle recommendation exploiting the Choquet integral. This allows us to formalize preferences for coalitions of environmental-related attributes, thus recommending product bundles accounting for synergies among product attributes. An experimental evaluation of a dataset of local food products in Northern Italy shows how the Choquet integral allows the natural formalization of a sensible notion of environmental friendliness and that standard approaches based on weighted sums of attributes end up recommending bundles with lower environmental friendliness even if weights are explicitly learned to maximize it. We further show how preference elicitation strategies can be leveraged to acquire weights of the Choquet integral from user feedback in terms of preferences over candidate bundles, and show how a handful of queries allow to recommend optimal bundles for a diverse set of user prototypes.

On restricted Falconer distance sets

Abstract We introduce a class of Falconer distance problems, which we call of restricted type, lying between the classical version and its pinned variant. Prototypical restricted distance sets are the diagonal distance sets, k-point configuration sets given by $$ \begin{align*}\Delta^{\mathrm{diag}}(E)= \{ \,|(x,x,\dots,x)-(y_1,y_2,\dots,y_{k-1})| : x, y_1, \dots,y_{k-1} \in E\, \}\end{align*} $$ for a compact $E\subset \mathbb {R}^d$ and $k\ge 3$ . We show that $\Delta ^{\mathrm{diag}}(E)$ has non-empty interior if the Hausdorff dimension of E satisfies (0.1) $$ \begin{align} \dim(E)> \begin{cases} \frac{2d+1}3, & k=3, \\ \frac{(k-1)d}k,& k\ge 4. \end{cases} \end{align} $$ We prove an extension of this to $C^\omega $ Riemannian metrics g close to the product of Euclidean metrics. For product metrics, this follows from known results on pinned distance sets, but to obtain a result for general perturbations g, we present a sequence of proofs of partial results, leading up to the proof of the full result, which is based on estimates for multilinear Fourier integral operators.

Uncertainty-weighted prototype active learning in domain adaptive semantic segmentation

Generalizing segmentation networks that adapt to unseen target domains is practical to the real-world utility. Recent self-training-based approaches compute the pseudo labels of highly imbalanced target image that are typically biased to the majority classes and basically noisy. To avoid the model falling into an error-prone and suboptimal state, we propose an uncertainty-weighted prototype active learning method to deal with domain adaptation regarding the semantic segmentation task. For assisting active learning to select reliable samples, multiple prototypical anchors strategy is employed to build the target image distribution model and assign accurate pixel-level pseudo labels, which consists of two different stages. In the initial training stage, this model defines each class as a set of non-learnable prototypes, and performs online clustering to identify target instances for annotating in the prototype-based feature space. In the fine-tuning stage, entropy criterion is adopted to generate representative class prototypes set of target domain, then compute the uncertainty as the clustering weight of prototype-level features via a predictive entropy-based strategy. By mixing these latent representations of the target samples and developing feature alignment loss function, our method can achieve more accurate segmentation performance. Extensive experiments on two benchmarks, GTAV → Cityscapes and SYNTHIA → Cityscapes, demonstrate that the proposed model outperforms other state-of-the-art approaches, and ablation study verifies the effectiveness of each key component. We hope our work will open a new view for future research in the field of domain adaptation. The code will be released soon at https://github.com/zihaodong/UPAL.

Climate Smart: geography, place and climate change adaptation education

ABSTRACT Geographical education provides a unique space for supporting climate literacy and action. In this article we present a novel place-based online educational resource platform – Climate Smart – which seeks to expand the capacity of young people to engage in one dimension of climate action: adaptation planning. The Climate Smart platform comprises a set of six online student workshops containing videos and quizzes and ends with a serious game: iAdapt. We outline the drivers and process behind the development of the platform, outlining the results of prototype testing and setting out how the platform is evolving. Positive results have been revealed through evaluative analysis and novel extensions of the platform are ongoing, although challenges remain regarding the longer-term sustainability of the platform.

REVIT SIMULATION APPLICATION IN EVALUATING FENESTRATION STRATEGIES TOWARDS ENERGY CONSERVATION IN MALAYSIAN INTERMEDIATE TERRACED HOUSES

With Malaysia’s population rising and rapid urbanisation, terraced houses were built to respond to new social and economic needs. The modern terraced house appeals to the demand for high-density development. However, with just two or three façades available for openings on the front and rear façades, these houses, particularly intermediate units, frequently lack natural ventilation, resulting in slow and inefficient air exchange throughout the house. As a result, adjustments on fenestration that functions as passive design solutions are necessary to increase internal thermal comfort and energy savings. This study aims to determine the various forms of fenestration placement in Malaysian terraced buildings and examine how these different types of fenestrations in terraced houses affect energy consumption and conservation. This study employs Revit modelling software to create a set of prototypes of intermediate terrace houses with different fenestration treatments. Building simulation is used to record findings of building energy performance based on the annual and monthly energy consumption of each prototype. This study found that the terraced home façade fenestration design significantly reduces energy consumption. It is anticipated that the finding of this study will benefit architects and designers better understand designing more ecologically conscious terraced houses by addressing internal thermal comfort through passive design.

AN ANALYSIS OF THE IMPLEMENTATION OF ACCESSIBILITY TOOLS ON WEBSITES

The websites of higher education institutions, due to the fact that they are addressed to multiple stakeholder groups, not only need to have an appropriately designed information structure but must also be useful. Additionally, in the case of public universities, their services are expected to be accessible to the widest possible audience, especially for people with disabilities. The accessibility tools used on websites should be quickly located, easily identifiable and user-friendly. So far, no standards have been developed regarding these issues, and therefore, there are various solutions on the web. The objective of this study is to analyze various implementations of accessibility tools on university websites in terms of their location, form of presentation and ways that enable access to them. A study was conducted in which web interfaces were evaluated with the participation of users. The experiment consisted of two parts: the first one used the eye tracking technique, whereas in the second one, a survey was conducted. The research material was prototypes of websites from four different universities. Each website had two versions differing in implementation of accessibility tools. In the study, 35 participants were divided into two groups of people. Each group was shown one of the two sets of website prototypes and the users were tasked with finding and activating a specific accessibility tool. After exploring the websites, each participant completed a questionnaire that pertained to their opinions regarding aspects such as appearance, placement and a way to access tools dedicated to people with disabilities. The obtained data, processed to the form of heatmaps and fixation maps, were subjected to a qualitative analysis. The survey results and eye tracking data were analyzed quantitatively. On the basis of performed analyzes it can be concluded that the following factors have an impact on the reduction in efficiency and productivity of users: placement of accessibility tools on university websites in a place other than the upper right corner, an indirect access to these tools or their non-standard appearance.

Categorization of Attitude to Reality in Paremiological Units

The study deals with the proverbal semantics from the point of view of its attitude to the surrounding world as an object of cognition, evaluation, verification and verbalization. The aim of the study is to describe the main semantic types of paremiological units (proverbs) according to the criterion of “relation to reality” in empirical, axiological, ontological and logical aspects. The research was done on the material of proverbs in the Kalmyk language from the collection of V.L. Kotvich (1905). The author’s methodology of categorizing the semantics of the aphoristic units, and their differentiation into semantic types according to the criterion of “relation to reality” was used. It has been revealed, that in the proverbs all major aspects of aphoristic categorization of the reality are represented, according to which the proverbial semantics is differentiated into six categories: “nomological” and “centential” (In the empirical aspect), “truistic” and “gregarious” (In the axiological aspect), and also paradoxical (In the logical aspect) and absurd (In the ontological aspect). It has been established that the categorization of the attitude to reality in proverbs is characterized by different quantitative representation and content heterogeneity. The attitude to reality as an object of cognition and evaluation is represented to the greatest extent in proverbs, and to the least extent - as an object of verification. Proverbs of different semantic categories have a specific set of prototypes, concepts and images. The categorisation of the attitude to reality in proverbs is universal, but it is distinguished by selectivity in terms of its distribution in the semantic types and conceptual imagery, which represents the ethnolinguistic and cultural specificity of the categorisation and conceptualisation of the proverb picture of the world.

Estimation of l0 norm penalized models: A statistical treatment

Fitting penalized models for the purpose of merging the estimation and model selection problem has become commonplace in statistical practice. Of the various regularization strategies that can be leveraged to this end, the use of the l0 norm to penalize parameter estimation poses the most daunting model fitting task. In fact, this particular strategy requires an end user to solve a non-convex NP-hard optimization problem irregardless of the underlying data model. For this reason, the use of the l0 norm as a regularization strategy has been woefully under utilized. To obviate this difficulty, a strategy to solve such problems that is generally accessible by the statistical community is developed. The approach can be adopted to solve l0 norm penalized problems across a very broad class of models, can be implemented using existing software, and is computationally efficient. The performance of the method is demonstrated through in-depth numerical experiments and through using it to analyze several prototypical data sets.

Characterization results of the first small pixel high rate (SPHIRD) pixel detector prototypes

This work presents a summary of the experimental results and the performance obtained from a first set of p-type silicon SPHIRD prototypes. The SPHIRD project (Small Pixel High Rate photon counting Detector) targets a new generation of X-ray photon counting hybrid pixel detectors for synchrotron radiation applications, with small pixels and operating between 10 and 30 keV. The readout ASIC is designed to explore techniques that boost the count rate capabilities of the detector and methods to manage, or in some cases exploit, the effects of charge sharing that are unavoidable with small pixels in this type of detector.The achievement of high photon rates relies on the implementation of a fast charge-sensitive amplifier in the pixelin combination with pile-up compensation techniques, which have been compared experimentally. The obtained results show and quantify their effectiveness in increasing the count-rate handled by the detector.The readout chip also includes dedicated circuitry for the relocation of photon hits, a functionality necessary to reduce the photon losses that are observed in photon counting detectors due to charge sharing when the discrimination threshold is set to 50% of the photon energy. The circuitry is however not limited to full pixel relocation: it also implements resources to reassign X-ray hits within regions smaller than the physical pixel. All these operation modes were experimentally evaluated with a pencil beam and with full field images.

Functional recovery evaluation of hybrid self‐centering piston‐based braced frames

AbstractSevere earthquakes possess formidable destructive potential, leading to substantial damage, widespread disruptions, and tragic loss of life. In response, researchers and policymakers strive to develop innovative solutions to quantify the seismic resilience of structures. This study aims to evaluate the post‐earthquake functional recovery of two types of braced frames: Self‐Centering Piston‐Based Braced Frames (SC‐PBBFs) and Buckling Restrained Braced Frames (BRBFs). The SC‐PBBFs are designed to mitigate structural damage by incorporating Self‐Centering Piston‐Based Bracings (SC‐PBBs) equipped with Shape Memory Alloy (SMA) bars and Friction Springs (FS). A set of building prototypes, representing low‐, mid‐, and high‐rise archetypes in seismic regions, are subjected to comprehensive analyses under 44 far‐field ground motions. This thorough examination incorporates seismic hazards, structural demands, and component damages to quantify the earthquake‐induced losses and repair timelines. The study findings indicate that BRBFs suffer more significant collapse losses than SC‐PBBFs. However, BRBF systems incur lower repair costs, mainly due to limited damage to acceleration‐sensitive non‐structural components. Additionally, the FEMA P‐58/ATC‐138 framework is employed to estimate re‐occupancy, functional recovery, and full repair time, identifying systems and components that exert a significant impact on functional recovery.

Prototype-oriented contrastive learning for semi-supervised medical image segmentation

Semi-supervised learning has been recently explored to advance medical image segmentation due to the challenges of acquiring sufficient labeled data. However, the mainstream semi-supervised learning methods leverage unlabeled data by enforcing the perturbation-based prediction consistency, which neglects the semantic relations across different images and cannot handle inductive bias problem about the feature distribution. To address the above problems, in this paper, a novel prototype-oriented contrastive learning framework is proposed for semi-supervised medical image segmentation. Specifically, inspired from prototypical learning, a set of prototypes is extracted first to represent the diverse feature distributions of different images. Then we present semi-supervised contrastive learning with a prototype-oriented sampling strategy, encouraging the network to explore voxel-wise semantic relations across different images and learn more discriminative features to enhance the segmentation ability. The effectiveness of the proposed method is illustrated on three public benchmark datasets. Extensive experiments show that the proposed method outperforms several state-of-the-art semi-supervised segmentation methods, demonstrating its effectiveness for the challenging semi-supervised medical image segmentation task.

Source-free domain adaptation with Class Prototype Discovery

Source-free domain adaptation requires no access to the source domain training data during unsupervised domain adaption. This is critical for meeting particular data sharing, privacy, and license constraints, whilst raising novel algorithmic challenges. Existing source-free domain adaptation methods rely on either generating pseudo samples/prototypes of source or target domain style, or simply leveraging pseudo-labels (self-training). They suffer from low-quality generated samples/prototypes or noisy pseudo-label target samples. In this work, we address both limitations by introducing a novel Class Prototype Discovery (CPD) method. In contrast to all alternatives, our CPD is established on a set of semantic class prototypes each constructed for representing a specific class. By designing a classification score based prototype learning mechanism, we reformulate the source-free domain adaptation problem to class prototype optimization using all the target domain training data, and without the need for data generation. Then, class prototypes are used to cluster target features to assign them pseudo-labels, which highly complements the conventional self-training strategy. Besides, a prototype regularization is introduced for exploiting well-established distribution alignment based on pseudo-labeled target samples and class prototypes. Along with theoretical analysis, we conduct extensive experiments on three standard benchmarks to validate the performance advantages of our CPD over the state-of-the-art models.

Prototype expansion and feature calibration for few-shot point cloud semantic segmentation

Few-shot point cloud semantic segmentation has attracted increasing attention due to its applicability to real-world scenarios in which each novel class contains only a few labelled data. Although existing methods have made great progress, a great challenge still remains: inaccurate decision boundaries are induced by the scarcity of labelled data. In this paper, a Prototype Expansion and Feature Calibration (PEFC) method is proposed, which fully exploits the information contained in query data to increase the amounts of class prototypes and calibrate these prototypes. To this end, a Prototype Expansion Module (PEM) is designed, which clusters query data into pseudo prototypes and expands the set of class prototypes with pseudo prototypes that have a high similarity to the class prototypes. Then, a Feature Calibration Module (FCM) is designed to calibrate these class prototypes, which is achieved by enabling bidirectional information flow between class prototypes and query features in a task-aware manner. Extensive experiments on two benchmark datasets show that the proposed PEFC outperforms the state-of-the-art methods by a large margin in all experimental settings.

The performance analysis of Hydraulic Ram Pump: Influence of specific parameters and validation with Comsol-Multiphysics

Both energy and environmental conservation have been major issues in today’s world. Conventional pumping devices use a huge amount of energy. This gave rise to interest in alternate pumping devices like hydraulic ram pumps. The hydraulic ram pump is capable of pumping water higher than its source with some inlet pressure and velocity head. It works on the principle of the water hammer effect which occurs due to sudden stoppage of flow resulting in the rise in pressure. Since it uses just two moving parts it is mechanically very simple, has very high reliability, minimal maintenance requirements, and long operational life. In rural areas where unconventional energy suffers from various limitations, emphasis is given mainly to the use of unconventional forms of energy. It is designed on the principle of conversion of the potential energy of water into kinetic energy and further conversion into pressure energy to lift water to some desired elevation without the use of electricity, motors, generators or batteries and that is the main advantage for opting this technology. It can operate autonomously, requiring minimal maintenance and having a long lifespan. Additionally, it is a green energy concept, environment-friendly, producing no emissions or pollutants. But, as it runs from the power of water itself in the form of waste water, efficiency is found to be minimal, rendering its limited scope. Hence, it is a need-based approach for maximizing its performance by providing a sustainable and efficient water pumping solution. The vital components of this technology have been detected which play a pivotal role in its performance. These are called influencing parameters. This paper investigates the effect of influencing parameters specifically–the horizontal distance between the pressure chamber and to waste valve (HD) on the performance and the overall efficiency of the Hydraulic ram pump (hydram). Both experimental and simulation studies were done in order to achieve the justification. Five sets of prototypes have been designed and tested individually for experimental and analytical studies. The maximum efficiency comes out to be 79 % with a flow rate of 140 L/hr for optimized HD of 17 cm. The results obtained from the experimental investigation are further simulated with comsol multiphysics for final validation. The experimental study and Comsol multiphysics simulation validate the proposed model and confirm its justification.

Affordable housing using industrialized construction: a market-based standardization of housing units in India

ABSTRACT Achieving affordable and mass-housing using industrialized construction needs robust space standardization of housing units. However, such standardization may fall short of market acceptance and result in a bad investment. Therefore, linking the market trend in developing space standardization is crucial, yet much less reported in the literature. This paper aims to develop market-oriented space standards for the mid-income group in India. The main purpose of developing affordable space standards is to create a set of modular design prototypes that can be translated into mass construction through a component-based open and Industrialized Building System (IBS). The paper uses a market study to assess the market trends of existing housing space standards in the National Capital Region around Delhi. Following the literature review, the paper sequentially presents the context, analysis, results, recommendations, and discussions. Dimensions for activity spaces of the housing module with attachment options are proposed in the range of 43.7 sq.m to 95.04 sq. m, including five unit prototypes. The industry will benefit from the developed design standards to manufacture large-scale housing. This paper provides a framework for developing similar space standards for other income groups in different contexts.

Cross-Domain Multi-Prototypes with Contradictory Structure Learning for Semi-Supervised Domain Adaptation Segmentation of Remote Sensing Images

Recently, unsupervised domain adaptation (UDA) segmentation of remote sensing images (RSIs) has attracted a lot of attention. However, the performance of such methods still lags far behind that of their supervised counterparts. To this end, this paper focuses on a more practical yet under-investigated problem, semi-supervised domain adaptation (SSDA) segmentation of RSIs, to effectively improve the segmentation results of targeted RSIs with a few labeled samples. First, differently from the existing single-prototype mode, a novel cross-domain multi-prototype constraint is proposed, to deal with large inter-domain discrepancies and intra-domain variations. Specifically, each class is represented as a set of prototypes, so that multiple sets of prototypes corresponding to different classes can better model complex inter-class differences, while different prototypes within the same class can better describe the rich intra-class relations. Meanwhile, the multi-prototypes are calculated and updated jointly using source and target samples, which can effectively promote the utilization and fusion of the feature information in different domains. Second, a contradictory structure learning mechanism is designed to further improve the domain alignment, with an enveloping form. Third, self-supervised learning is adopted, to increase the number of target samples involved in prototype updating and domain adaptation training. Extensive experiments verified the effectiveness of the proposed method for two aspects: (1) Compared with the existing SSDA methods, the proposed method could effectively improve the segmentation performance by at least 7.38%, 4.80%, and 2.33% on the Vaihingen, Potsdam, and Urban datasets, respectively; (2) with only five labeled target samples available, the proposed method could significantly narrow the gap with its supervised counterparts, which was reduced to at least 4.04%, 6.04%, and 2.41% for the three RSIs.