The achievement of robot autonomy has environmental perception as a prerequisite. The hazards rendered from uneven, soft and slippery terrains, which are generally named non-geometric hazards, are another potential threat reducing the traversing efficient, and therefore receiving more and more attention from the robotics community. In the paper, the vibration-based terrain classification (VTC) is investigated by taking a very practical issue, i.e., lack of labels, into consideration. According to the intrinsic temporal correlation existing in the sampled terrain sequence, a modified Laplacian SVM is proposed to utilise the unlabelled data to improve the classification performance. To the best of our knowledge, this is the first paper studying semi-supervised learning problem in robotic terrain classification. The experiment demonstrates that: (1) supervised learning (SVM) achieves a relatively low classification accuracy if given insufficient labels; (2) feature-space homogeneity based semi-supervised learning (traditional Laplacian SVM) cannot improve supervised learning’s accuracy, and even makes it worse; (3) feature- and temporal-space based semi-supervised learning (modified Laplacian SVM), which is proposed in the paper, could increase the classification accuracy very significantly.
This paper re-examines from an evolutionary perspective the typological status of Chinese, with regard to the issue of how the information of motion events is encoded (Talmy 2000;Slobin 2004). We investigate, with emphasis on the roles of both language structure and language use, the four periods of Chinese (Old, Middle, Pre-Modern and Modern) in terms of parameters such as path, manner and ground, and compare with typologically different languages, namely, verb-framed languages like Spanish and satellite-framed languages like English. Our statistical study shows that (i) Chinese has been undergoing a typological shift from a verb-framed language to a satellite-framed language, and Pre-Modern Chinese is a stepped-up period with respect to the speed of evolution; (ii) Modern Chinese adopts diverse patterns to encode motion events, which are different from both typical verb-framed languages and typical satellite-framed languages. We thus conclude that (i) contrary to Peyraube's (2006) claim, the typological shift in Chinese has not yet been achieved; (ii) there is little justification for classifying Chinese as an equipollently framed language as in Slobin (2004) and Chen and Guo (2009). Therefore, there is no need to posit an equipollent type for Chinese; and (iii) Modern Chinese is in a transitional state from a V-type to an S-type due to the evolution of its motion expressions. It also favors the S-type in both its morphosyntactic properties and language use.
This paper reports on a corpus-based study aimed at reexamining the typological status and diachronic change of motion expressions in Chinese, drawing on parallel texts consisting of autonomous motion expressions in Old Chinese (OC) and its Modern Chinese (MoC) translation. The results show that MoC significantly differs from OC both in the preference of lexicalization patterns (Talmyan typology) and semantic components distributed in discourse (Slobinian typology) when narrating similar motion scenes. However, these results fail to support the viewpoint that Chinese has undergone a change from a verb- to a satellite-frame (Li 1993; Talmy 2000; Peyraube 2006; Shi & Wu 2014). It is argued that (i) the Talmyan typology and the Slobinian typology should be treated separately. In Talmyan typology, the diachrony of Chinese demonstrates the change of a V- to a parallel-frame, in that satellite- and verb-framed constructions in MoC have equal frequency and show no bias for the encoding of subtypes of autonomous motion. In Slobinian typology, MoC remains as a Path-salient language, as it gives considerable weight to the expression of Path; (ii) the dominant lexicalization pattern in a language varies from one sub-domain of motion to another (see also Lamarre 2003), and thus the typology of motion expressions is sub-domain-specific; and (iii) motivating forces and blocking forces, furthermore, co-exist diachronically for the typological evolution of motion encoding due to the idiosyncrasy of the morphosyntactic system.
The competition among enterprises is becoming increasingly fierce. The research on the financial management evaluation model is helpful for enterprises to find possible risks as soon as possible. This paper constructs the financial management evaluation model based on the deep belief network and applies the analytic hierarchy process to determine the weight of financial management evaluation indicators, which is compared with other classical deep learning evaluation methods, such as KNN, SVM-RBF, and SVM linear. It has achieved an accuracy of more than 81%, showing a satisfactory prediction effect, which is of great significance to formulate corresponding countermeasures, strengthen financial management, improve the capital market system, and promote high-quality economic development. In addition, aiming at the problem of abnormal financial data, this paper uses the new sample dataset obtained by principal component analysis for convolution neural network model learning, which enhances the prediction accuracy of the model and fully shows that deep learning is feasible in the research of financial management prediction, and there is still a lot of space to explore.
Physics-Informed Neural Network (PINN) has become a commonly used machine learning approach to solve partial differential equations (PDE). But, facing highdimensional second-order PDE problems, PINN will suffer from severe scalability issues since its loss includes second-order derivatives, the computational cost of which will grow along with the dimension during stacked back-propagation. In this paper, we develop a novel approach that can significantly accelerate the training of Physics-Informed Neural Networks. In particular, we parameterize the PDE solution by the Gaussian smoothed model and show that, derived from Stein's Identity, the second-order derivatives can be efficiently calculated without backpropagation. We further discuss the model capacity and provide variance reduction methods to address key limitations in the derivative estimation. Experimental results show that our proposed method can achieve competitive error compared to standard PINN training but is two orders of magnitude faster.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.