Transferring Annotator- and Instance-dependent Transition Matrix for Learning from Crowds.

Learning from crowds describes that the annotations of training data are obtained with crowd-sourcing services. Multiple annotators each complete their own small part of the annotations, where labeling mistakes that depend on annotators occur frequently. Modeling the label-noise generation process by the noise transition matrix is a powerful tool to tackle the label noise. In real-world crowd-sourcing scenarios, noise transition matrices are both annotator- and instance-dependent. However, due to the high complexity of annotator- and instance-dependent transition matrices (AIDTM), annotation sparsity, which means each annotator only labels a tiny part of instances, makes modeling AIDTM very challenging. Without prior knowledge, existing works simplify the problem by assuming the transition matrix is instance-independent or using simple parametric ways, which lose modeling generality. Motivated by this, we target a more realistic problem, estimating general AIDTM in practice. Without losing modeling generality, we parameterize AIDTM with deep neural networks. To alleviate the modeling challenge, we suppose every annotator shares its noise pattern with similar annotators, and estimate AIDTM via knowledge transfer. We hence first model the mixture of noise patterns by all annotators, and then transfer this modeling to individual annotators. Furthermore, considering that the transfer from the mixture of noise patterns to individuals may cause two annotators with highly different noise generations to perturb each other, we employ the knowledge transfer between identified neighboring annotators to calibrate the modeling. Theoretical analyses are derived to demonstrate that both the knowledge transfer from global to individuals and the knowledge transfer between neighboring individuals can effectively help mitigate the challenge of modeling general AIDTM. Experiments confirm the superiority of the proposed approach on synthetic and real-world crowd-sourcing data. The implementation is available at https://github.com/tmllab/TAIDTM.

Multimodal Composition Example Mining for Composed Query Image Retrieval.

Composed query image retrieval task aims to retrieve the target image in the database by a query that composes two different modalities: a reference image and a sentence declaring that some details of the reference image need to be modified and replaced by new elements. Tackling this task needs to learn a multimodal embedding space, which can make semantically similar targets and queries close but dissimilar targets and queries as far away as possible. Most of the existing methods start from the perspective of model structure and design some clever interactive modules to promote the better fusion and embedding of different modalities. However, their learning objectives use conventional query-level examples as negatives while neglecting the composed query's multimodal characteristics, leading to the inadequate utilization of the training data and suboptimal construction of metric space. To this end, in this paper, we propose to improve the learning objective by constructing and mining hard negative examples from the perspective of multimodal fusion. Specifically, we compose the reference image and its logically unpaired sentences rather than paired ones to create component-level negative examples to better use data and enhance the optimization of metric space. In addition, we further propose a new sentence augmentation method to generate more indistinguishable multimodal negative examples from the element level and help the model learn a better metric space. Massive comparison experiments on four real-world datasets confirm the effectiveness of the proposed method.

Dopamine-Mediated Graphene Bridging Hexagonal Boron Nitride for Large-Scale Composite Films with Enhanced Thermal Conductivity and Electrical Insulation.

Heat accumulation generated from confined space poses a threat to the service reliability and lifetime of electronic devices. To quickly remove the excess heat from the hot spot, it is highly desirable to enhance the heat dissipation in a specific direction. Herein, we report a facile route to fabricate the large-scale composite film with enhanced thermal conductivity and electrical insulation. The well-stacked composite films were constructed by the assembly of polydopamine (PDA)-modified graphene nanosheets (GNSPDA) and hexagonal boron nitride (BNPDA), as well as bacterial cellulose (BC). The introduction of the PDA layer greatly improves the interface compatibility between hybrid fillers and BC matrix, and the presence of GNSPDA-bridging significantly increases the probability of effective contact with BNPDA fillers, which is beneficial to form a denser and complete "BN-GNS-BN" heat conduction pathway and tight filler-matrix network, as supported by the Foygel model fitting and numerical simulation. The resulting BC/BNPDA/GNSPDA film shows the thermal conductivity and tensile strength of 34.9 W·m-1·K-1 and 30.9 MPa, which separately increases to 161% and 155% relative to the BC/BNPDA film. It was found that the low electrically conductive and high thermal conductive properties can be well balanced by tuning the mass ratio of GNSPDA at 5 wt%, and the electrical conductivity caused by GNSPDA can be effectively blocked by the BNPDA filler network, giving the low electrical conductivity of 1.8 × 10-10 S·cm-1. Meanwhile, the BC/BNPDA/GNSPDA composite films effectively transfer the heat and diminish the hot-spot temperature in cooling LED chip module application. Thus, the present study may pave the way to promoting the industrialization of scalable thermal management devices.

Case Report: Neonatal genital prolapse in a pair of premature Twins.

Neonatal genital prolapse is a rare situation. This report presents a unique case involving a pair of premature female twins who both developed vaginal wall prolapse without any neurological deficits. Multiple factors such as selective intrauterine growth restriction, feeding intolerance, extrauterine growth retardation, and elevated intra-abdominal pressure after birth may have contributed to the development of this phenomenon. Notably, the severity of prolapse was more pronounced in the twin with lower birth weight and smaller for gestational age. After a five-month follow-up period, the twins' prolapsed vaginal wall fully retracted due to a combination of conservative treatment and enhanced nutritional support.

Catalytic activity of 1D chains of gold oxide on a stepped gold surface from density functional theory.

The rich surface chemistry of gold at the nanoscale has made it an important catalyst for low-temperature applications. Recent studies point to the possible role of self-organized structures formed by chemisorbed O atoms on the surface of gold catalysts for their catalytic activity and/or deactivation. In this study, we investigate the reactivity of a double O chain running along a step on a Au(221) surface with oxygen vacancies as a prototypical model of a 1D surface gold oxide. We compare CO and O2 adsorption on such a chain with the oxygen-free Au(221) surface model. A systematic study of the reactivity of the double chain with O vacancies was done with respect to the regular Au(221) surface using CO as a probe. The CO oxidation was investigated assuming dissociative and associative mechanisms. Remarkably, O2 adsorbs stronger on the double oxygen vacancy than on the regular Au(221) surface, and its dissociation barrier reduces significantly from 1.84 eV to 0.87 eV, whereas the CO adsorption energy is similar on these surfaces. Calculations suggest that CO oxidation should occur more efficiently on the double O vacancy than on the regular Au(221) surface due to stronger adsorption of O2 and a low activation barrier for O2 + CO surface reaction.

Risk and protection strategies of Amolops wuyiensis intestine against gastrointestinal nematode (Cosmocercoides wuyiensis n. sp.) infection.

Anuran amphibians are susceptible to infection by intestinal nematodes, but the damage and response mechanisms that occur in their intestines after infection are only partially understood. In this study, the intestinal disruption and response mechanisms in Amolops wuyiensis frogs infected with Cosmocercoides wuyiensis n. sp. were revealed through analysis of the intestinal histopathology, digestive enzyme activity, transcriptome and intestinal microbiota. Tissue section analysis showed histological damage and inflammation in the infected intestine, and the digestive enzyme activity indicated a decrease in digestion and absorption of some nutrients. We found that infection led to differences in the intestinal microbiota composition, including lower diversity and symbiotic relationships. The greater relative abundance of the genera Burkholderia and Rhodococcus may enhance intestinal immunity to resist pathogenic infections. A comparison of the transcriptomes of infected and uninfected intestines revealed 1055 differentially expressed genes. GO enrichment and KEGG pathways analyses suggested that the guts of infected C. wuyiensis n. sp. show enhanced complement activation, cell adhesion molecule function, NOD-like receptor signalling pathway activity and other innate immunity responses. Among the adaptive immune responses, the intestinal immune network for IgA production was significantly enriched, and the expression of IL-17D and transforming growth factor beta-1 genes were upregulated in the infected intestine. These results imply that C. wuyiensis n. sp. infection of A. wuyiensis intestine may trigger innate and adaptive immune responses, which reduce the post-infection burden. Furthermore, the intestine of A. wuyiensis may also respond to C. wuyiensis n. sp. infection by increasing metallocarboxypeptidase activity and accelerating smooth muscle contraction.

Assembly of Hexagonal Column Interpenetrated Spheres from Plant Polyphenol/Cationic Surfactants and Their Application as Antimicrobial Molecular Banks.

Microbial infections have become a great threat to human health and one of the main risks arises from direct contact with the surfaces contaminated by pathogenic microbes. Herein, a kind of hexagonal column interpenetrated spheres (HCISs) are fabricated by non-covalent assembly of plant gallic acid with quaternary ammonium surfactants. Different from one-time burst release of conventional antimicrobial agents, the HCIS acts like a "antimicrobial molecular bank" and releases the antimicrobial ingredients in a multistage way, leading to long-lasting antimicrobial performance. Taking advantage of strong hydrophobicity and adhesion, HCISs are applicable to various substrates and endowed with anti-water washing property, thus showing high in vitro antimicrobial efficiency (>99 %) even after being used for 10 cycles. Meanwhile, HCISs exhibit broad-spectrum antimicrobial activity against bacteria and fungi, and have good biocompatibility with mammalian cells. Such a low-cost and portable long-lasting antimicrobial agent meets the growing anti-infection demand in public spaces.

Magnetic Resonance Imaging Features of Cerebral Infarction in Critical Patients Based on Convolutional Neural Network.

The clinical application of the artificial intelligence-assisted system in imaging was investigated by analyzing the magnetic resonance imaging (MRI) influence characteristics of cerebral infarction in critically ill patients based on the convolutional neural network (CNN). Fifty patients with cerebral infarction were enrolled and examined by MRI. Besides, a CNN artificial intelligence system was established for learning and training. The features were extracted from the MRI image results of the patients, and then, the data were calculated by computer technology. The gray-level cooccurrence matrix (GLCM) of T1-weighted images was 0.872 ± 0.069; the reasonable prediction (ALL) result was 0.766 ± 0.112; the gray-level run-length matrix (GLRLM) was 0.812 ± 0.101; the multigray-level area size matrix (MGLSZM) result was 0.713 ± 0.104; and the result of gray-scale area size matrix (GLSZM) was 0.598 ± 0.099. The GLCM, ALL, GLRLM, MGLSZM, and GLSZM of enhanced T1-weighted images were 0.710 ± 0.169, 0.742 ± 0.099, 0.778 ± 0.096, 0.801 ± 0.104, and 0.598 ± 0.099, respectively. The GLCM, ALL, GLRLM, MGLSZM, and GLSZM of T2-weighted images were 0.780 ± 0.096, 0.798 ± 0.087, 0.888 ± 0.086, 0.768 ± 0.112, and 0.767 ± 0.100, respectively. In short, the image diagnosis method that could reduce the subjective visual judgment error to a certain extent was found by analyzing the characteristics of MRI images of critically ill patients with cerebral infarction based on CNN.

Dynamic transcriptome and histomorphology analysis of developmental traits of hindlimb thigh muscle from Odorrana tormota and its adaptability to different life history stages.

BACKGROUND: Systematic studies on the development and adaptation of hindlimb muscles in anura amphibians are rare. Here, we integrated analysis of transcriptome and histomorphological data for the hindlimb thigh muscle of Odorrana tormota (concave-eared torrent frog) at different developmental stages, to uncover the developmental traits of hindlimb thigh muscle from O. tormota and its adaptability to different life history stages. RESULTS: The development of hindlimb thigh muscle from O. tormota has the following characteristics. Before metamorphosis, myogenous cells proliferate and differentiate into myotubes, and form 11 muscle groups at G41; Primary myofibers and secondary myofibers appeared during metamorphosis; 11 muscle groups differentiated continuously to form myofibers, accompanied by myofibers hypertrophy after metamorphosis; During the growth process of O. tormota from G42 to G46, there were differences between the sexes in the muscle groups that differentiate into muscle fibers, indicating that there was sexual dimorphism in the hindlimb thigh muscles of O. tormota at the metamorphosis stages. Some genes and pathways related to growth, development, and movement ability of O. tormota at different developmental stages were obtained. In addition, some pathways associated with adaptation to metamorphosis and hibernation also were enriched. Furthermore, integrated analysis of the number of myofibers and transcriptome data suggested that myofibers of specific muscle groups in the hindlimbs may be degraded through lysosome and ubiquitin pathways to transform into energy metabolism and other energy-related substances to meet the physiological needs of hibernation. CONCLUSIONS: These results provide further understanding the hindlimb thigh muscle development pattern of frogs and their adaption to life history stages.

Manipulation of composting oxygen supply to facilitate dissolved organic matter (DOM) accumulation which can enhance maize growth.

Promotion of crop yield by compost application is generally thought to be ascribed to a better supply of macro and micronutrients, however the importance of compost DOM on plant growth has not been well demonstrated. In this study, composting of chicken manure, spent mushroom and sawdust was conducted under aerobic or anaerobic condition to determine the effects of compost DOM on plant growth. It was found that dissolved organic matter (DOM) first increased and then decreased in compost, and DOM of anaerobic compost was slightly higher than that of aerobic compost. When compost extract was applied to maize, among N, P, K and DOM content, it was DOM content that was most significantly and strongly related to plant biomass (r = 0.843, p＜0.001). Compost DOM was also strongly related to soil properties, the improvement of which can also promote plant growth. Compost application confirmed that higher compost DOM results in greater plant biomass. In order to facilitate compost DOM accumulation, we designed a novel composting process which combined aerobic and anaerobic treatments, and the resulting compost (A-Ana compost) with the highest amount of DOM displayed the best performance in promotion of plant growth. A-Ana compost was able to increase maize biomass by 32.71% and 12.40% compared with only anaerobic or aerobic compost, respectively. Therefore, DOM is a critical factor determining compost quality and it is feasible to manipulate composting oxygen supply condition to increase compost DOM, which will lead to increased plant yield.

Cascaded Correlation Refinement for Robust Deep Tracking.

Recent deep trackers have shown superior performance in visual tracking. In this article, we propose a cascaded correlation refinement approach to facilitate the robustness of deep tracking. The core idea is to address accurate target localization and reliable model update in a collaborative way. To this end, our approach cascades multiple stages of correlation refinement to progressively refine target localization. Thus, the localized object could be used to learn an accurate on-the-fly model for improving the reliability of model update. Meanwhile, we introduce an explicit measure to identify the tracking failure and then leverage a simple yet effective look-back scheme to adaptively incorporate the initial model and on-the-fly model to update the tracking model. As a result, the tracking model can be used to localize the target more accurately. Extensive experiments on OTB2013, OTB2015, VOT2016, VOT2018, UAV123, and GOT-10k demonstrate that the proposed tracker achieves the best robustness against the state of the arts.

Assessment of PBE+U and HSE06 methods and determination of optimal parameter U for the structural and energetic properties of rare earth oxides.

Rare earth oxides are attracting increasing interest as a relatively unexplored group of materials with potential applications in heterogeneous catalysis and electrocatalysis; therefore, a credible and universal computational approach is needed for modeling their reactivity. In this work, we systematically assessed the performance of the PBE+U method against the results of the hybrid HSE06 method with respect to the description of structural parameters and energetic properties of the selected hexagonal lanthanide sesquioxides and the cubic fluorite-type cerium dioxide. In addition, we evaluated the performance of PBE+U in describing the electronic structure and adsorption properties of the CeO2(111) and Nd2O3(0001) surfaces. The HSE06 method reproduces rather well the lattice parameters and selected energetic properties with respect to the experimental values. The PBE+U method is able to reproduce the results of HSE06 or the experimental values only if the U parameter is selected from an appropriate range of values. The U value around 3 eV gives the best description of the lattice parameters of most bulk oxides. 2 eV-3 eV is also found to be the optimal range of U for the reaction energies of bulk La2O3, Ce2O3, Nd2O3, Er2O3, and Ho2O3. U = 1 eV gives the best results for Pr2O3, Pm2O3, Eu2O3, Tm2O3, and Lu2O3, whereas Gd2O3 could not be accurately described by the PBE+U method. The U values (∼3 eV) found optimal for most bulk oxides also work well in the calculations of adsorption of small molecules on Nd2O3(0001) and CeO2(111), although larger U values are required to obtain sufficient localization of 4f electrons.

Integrated analysis of mRNA and miRNA expression profiles reveals muscle growth differences between fast- and slow-growing king ratsnakes (Elaphe carinata).

In some countries, snakes are important protein sources in human diets, and their economic value depends predominantly on their muscle production, including in the king ratsnake (Elaphe carinata). Muscle growth in the king ratsnake clearly differs among individuals. To date, few potential molecular mechanisms underlying these differences in muscle growth and development have been reported. Here, we integrated mRNA and miRNA expression profiles to screen for genes, pathways, and predicted miRNA-mRNA networks associated with muscle growth and development in fast-growing and slow-growing King ratsnakes. Six hundred eight differentially expressed genes (DEGs) were identified, 48 of which were associated with muscle growth. The 37 genes upregulated in fast-growing individuals (FGIs) may be related to the promotion of muscle growth, whereas the 11 upregulated genes in slow-growing individuals (SGIs) may be related to the inhibition of muscle growth. Seven DEGs were enriched in the PI3K-AKT-MTOR signaling pathway, which appears to promote muscle growth in FGIs. Eleven DEGs were enriched in the ubiquitin-proteasome pathway, which appears to inhibit muscle growth in SGIs. It may interpret why muscle growth differences. Furthermore, 698 miRNA were identified, including 125 novel miRNAs. 63 differentially expressed miRNA (DEMs) were screened, and 950 negative miRNA-mRNA interactions with the 63 DEMs and 608 DEGs were predicted. The miRNA-targeted genes were enriched in pathways related to muscle growth, protein synthesis, and protein degradation. Therefore, in addition to the identified DEGs, miRNAs may play important roles in the differential regulation of muscle growth in FGIs and SGIs of the king ratsnake.

MXene Surface Terminations Enable Strong Metal-Support Interactions for Efficient Methanol Oxidation on Palladium.

Efficient catalysis of the methanol oxidation reaction (MOR) greatly determines the widespread implementation of direct methanol fuel cells. Exploring a suitable support for noble metal catalysts with regard to decreasing the mass loading and optimizing the MOR activity remains a key challenge. Herein, we achieve an over 60% activity enhancement of a palladium (Pd) catalyst by introducing a two-dimensional Ti3C2Tx MXene as the support compared to a commercial Pd/C catalyst. Not only are more catalytically active Pd sites exposed on the Pd/MXene catalyst while maintaining a low mass loading, but the introduction of the MXene support also significantly alters the surface electronic structure of Pd. Specifically, spectroscopy and density functional theory (DFT) computations indicate that sufficiently electronegative terminations of the Ti3C2Tx MXene surface can induce strong metal-support interactions (SMSI) with the Pd catalyst, leading to optimal methanol adsorption. This MXene-supported Pd catalyst exhibits a much higher MOR current density (12.4 mA cm-2) than that of commercial Pd/C (7.6 mA cm-2). Our work largely optimizes the intrinsic activity of a Pd catalyst by the utilization of MXene surface terminations, and the crucial SMSI effects revealed herein open a rational avenue to the design of more efficient noble metal catalysts for MOR.

Tracking down the origin of peculiar vibrational spectra of aromatic self-assembled thiolate monolayers.

Several studies have previously observed surprisingly low frequencies for the C-H stretching modes of self-assembled monolayers (SAMs) prepared from aromatic thiols. The reason for this property has so far remained elusive. Therefore, we report a novel study of the vibrational spectra of SAMs prepared on Au from two different aromatic thiols, namely, 4'-nitro-1,1'-biphenyl-4-thiol (NBPT) and 4-aminothiophenol (ATP). The SAMs were prepared by vapor deposition (VD) in ultrahigh vacuum (UHV) as well as by the solution method (SM) and their quality was controlled by X-ray photoelectron spectroscopy (XPS). In addition, amino terminated SAMs were also obtained by electron irradiation and by chemical reduction of NBPT SAMs. Beside infrared reflection absorption spectroscopy (IRRAS), we have employed high resolution electron energy loss spectroscopy (HREELS), by which VD SAMs can be studied in situ, i.e. without exposing them to air. Hence, we can exclude possible contributions of solvent molecules to the vibrational spectra. Nonetheless, HREELS in fact reveals the same large red shift of the C-H stretching modes in the SAMs as also observed in ex situ IRRAS experiments. In contrast, HREELS for physisorbed ATP and ATP in a KBr pellet measured by transmission infrared spectroscopy exhibit the expected aromatic bands. Using a computational approach, we can exclude molecular packing effects as origin of this shift. Therefore, we propose chemical changes in the aromatic rings during SAM formation as an alternative explanation for the observed frequency shift. As another striking effect, the N-H stretching vibrational modes of the amino-terminated SAMs are extremely weak in both IRRAS and HREELS despite the fact that XPS confirms the presence of amino groups. A very weak signal is observed only in the case of an electron irradiated NBPT SAM. In contrast, an energy loss ascribed to the N-H stretching vibrations is clearly observed in HREELS of ATP physisorbed on an ATP SAM and on graphite as well as in the transmission infrared spectrum of ATP in KBr. The extremely low intensity of these vibrations in the SAM is traced back to the inherently low transition dipole moment for the excitation of N-H stretching modes in free N-H groups. Furthermore, the calculations suggest that the much stronger signals of N-H stretching modes involved in hydrogen-bonding with adjacent amino groups are suppressed because these vibrations are oriented parallel to the surface.

Excellent Performances of Dealuminated H-Beta Zeolites from Organotemplate-Free Synthesis in Conversion of Biomass-derived 2,5-Dimethylfuran to Renewable p-Xylene.

Direct synthesis of renewable p-xylene (PX) by cycloaddition of biomass-derived 2,5-dimethylfuran (2,5-DMF) and ethylene was achieved over Al-rich H-beta zeolites synthesized by an organotemplate-free approach and their dealuminated counterparts with different Si/Al ratios. Among them, H-beta zeolite with an Si/Al ratio of 22, obtained from an Al-rich parent by dealumination, was found to be an excellent catalyst for the synthesis of PX. A PX yield of 97 % and 2,5-DMF conversion of 99 % were obtained under optimized conditions. These results are even better than those of a commercial H-beta zeolite prepared using a organotemplate synthesis with a similar Si/Al ratio of 19. The excellent performance of the H-beta zeolite with Si/Al ratio of 22 is closely related to its acidity and porous structure. A moderate Brønsted/Lewis acid ratio can improve the conversion of 2,5-DMF to as high as 99 %. Furthermore, dealuminated H-beta zeolite has a secondary pore system that facilitates product diffusion, which increases the selectivity to PX. In addition, this catalyst shows better regeneration. After five successive regeneration cycles, the yield of PX was still as high as 85 % without obvious dealumination. This work provides a deeper understanding of the more general Diels-Alder cycloaddition of furan-based feedstocks and olefins and significantly improves the potential for the synthesis of chemicals from lignocellulosic biomass.

The research topic landscape in the literature of social class and inequality.

The literature of social class and inequality is not only diverse and rich in sight, but also complex and fragmented in structure. This article seeks to map the topic landscape of the field and identify salient development trajectories over time. We apply the Latent Dirichlet Allocation topic modeling technique to extract 25 distinct topics from 14,038 SSCI articles published between 1956 to 2017. We classified three topics as "hot", eight as "stable" and 14 as "cold", based on each topic's idiosyncratic temporal trajectory. We also listed the three most cited references and the three most popular journal outlets per topic. Our research suggests that future effort may be devoted to Topics "urban inequalities, corporate social responsibility and public policy in connected capitalism", "education and social inequality", "community health intervention and social inequality in multicultural contexts" and "income inequality, labor market reform and industrial relations".

Insights into the Correlation of Aluminum Distribution and Brönsted Acidity in H-Beta Zeolites from Solid-State NMR Spectroscopy and DFT Calculations.

Here we utilized 27Al MAS/MQMAS and 31P MAS NMR of quantitative adsorption of trimethylphosphine oxide (TMPO) and DFT calculations to elucidate the relationship between Al distribution and Brönsted acidity of series H-Beta zeolites derived from dealumination of Al-rich H-Beta zeolite. Three types of Brönsted acid strengths corresponding to different specific Al T-sites were demonstrated. The removal of one framework Al in 5MR2--2Al and 6MR-2Al sites led to increasing the Brönsted acid strength of dealuminated H-Beta. Our findings on such exact correlation between specific Al distributions and corresponding Brönsted acid sites may guide the controlling Al distribution to get desired acid properties through zeolite synthesis or finely tuned dealumination, which has a great impact on the catalytic activity and selectivity of zeolite catalysts.

Theoretical and experimental study on a compliant flipper-leg during terrestrial locomotion.

An amphibious robot with straight compliant flipper-legs can conquer various amphibious environments. The robot can rotate its flipper-legs and utilize their large deflection to walk on rough terrain, and it can oscillate the straight flipper-legs to propel itself underwater. This paper focuses on the dynamics of the compliant straight flipper-legs during terrestrial locomotion by modeling its deformation dynamically with large deflection theory and simulating it to investigate the parameters of locomotion such as trajectory, velocity, and propulsion. To validate the theoretical model of dynamic locomotion, a single-leg experimental platform is used to explore the flipper-legs in motion with various structural and kinematic parameters. Furthermore, a robotic platform mounting with four compliant flipper-legs is also developed and used to experiment with locomotion. The trajectories of the rotating axle of the compliant flipper-leg during locomotion were approximately coincidental in simulation and in experiments. The speed of locomotion and cost of transport during locomotion were explored and analyzed. The performance of different types of compliant flipper-legs during locomotion shows that varying the degrees of stiffness will have a significant effect on their locomotion. The dynamic model and analysis of the compliant flipper-leg for terrestrial locomotion facilitates the ability of amphibious robots to conquer complex environments.

2-(2-Hy-droxy-2-phenyleth-yl)-1-methyl-cyclo-propan-1-ol.

The asymmetric unit of the title compound, C(12)H(16)O(2), contains two independent mol-ecules in which the dihedral angles between the benzene and cyclo-propane rings are 75.9 (3) and 76.3 (3)°. In the crystal, the mol-ecules are connected by O-H⋯O hydrogen bonds into a three dimensional supra-molecular structure.