The use of language is subject to variation over time as well as across social groups and knowledge domains, leading to differences even in the monolingual scenario. Such variation in word usage is often called lexical semantic change (LSC). The goal of LSC is to characterize and quantify language variations with respect to word meaning, to measure how distinct two language sources are (that is, people or language models). Because there is hardly any data available for such a task, most solutions involve unsupervised methods to align two embeddings and predict semantic change with respect to a distance measure. To that end, we propose a self-supervised approach to model lexical semantic change by generating training samples by introducing perturbations of word vectors in the input corpora. We show that our method can be used for the detection of semantic change with any alignment method. Furthermore, it can be used to choose the landmark words to use in alignment and can lead to substantial improvements over the existing techniques for alignment. We illustrate the utility of our techniques using experimental results on three different datasets, involving words with the same or different meanings. Our methods not only provide significant improvements but also can lead to novel findings for the LSC problem.
This research investigates the Statistical Machine Translation approaches to translate speech in real time automatically. Such systems can be used in a pipeline with speech recognition and synthesis software in order to produce a real-time voice communication system between foreigners. We obtained three main data sets from spoken proceedings that represent three different types of human speech. TED, Europarl, and OPUS parallel text corpora were used as the basis for training of language models, for developmental tuning and testing of the translation system. We also conducted experiments involving part of speech tagging, compound splitting, linear language model interpolation, TrueCasing and morphosyntactic analysis. We evaluated the effects of variety of data preparations on the translation results using the BLEU, NIST, METEOR and TER metrics and tried to give answer which metric is most suitable for PL-EN language pair.
In recent years, Convolutional Neural Networks (CNNs) have enabled ubiquitous image processing applications. As such, CNNs require fast runtime (forward propagation) to process high-resolution visual streams in real time. This is still a challenging task even with state-of-the-art graphics and tensor processing units. The bottleneck in computational efficiency primarily occurs in the convolutional layers. Performing operations in the Fourier domain is a promising way to accelerate forward propagation since it transforms convolutions into elementwise multiplications, which are considerably faster to compute for large kernels. Furthermore, such computation could be implemented using an optical 4f system with orders of magnitude faster operation. However, a major challenge in using this spectral approach, as well as in an optical implementation of CNNs, is the inclusion of a nonlinearity between each convolutional layer, without which CNN performance drops dramatically. Here, we propose a Spectral CNN Linear Counterpart (SCLC) network architecture and develop a Knowledge Distillation (KD) approach to circumvent the need for a nonlinearity and successfully train such networks. While the KD approach is known in machine learning as an effective process for network pruning, we adapt the approach to transfer the knowledge from a nonlinear network (teacher) to a linear counterpart (student). We show that the KD approach can achieve performance that easily surpasses the standard linear version of a CNN and could approach the performance of the nonlinear network. Our simulations show that the possibility of increasing the resolution of the input image allows our proposed 4f optical linear network to perform more efficiently than a nonlinear network with the same accuracy on two fundamental image processing tasks: (i) object classification and (ii) semantic segmentation.
Facing the rapid growth in the issuance of financial tickets (or bills, invoices etc.), traditional manual invoice reimbursement and financial accounting system are imposing an increasing burden on financial accountants and consuming excessive manpower. To solve this problem, we proposes an iterative self-learning Framework of Financial Ticket intelligent Recognition System (FFTRS), which can support the fast iterative updating and extensibility of the algorithm model, which are the fundamental requirements for a practical financial accounting system. In addition, we designed a simple yet efficient Financial Ticket Faster Detection network (FTFDNet) and an intelligent data warehouse of financial ticket are designed to strengthen its efficiency and performance. At present, the system can recognize 194 kinds of financial tickets and has an automatic iterative optimization mechanism, which means, with the increase of application time, the types of tickets supported by the system will continue to increase, and the accuracy of recognition will continue to improve. Experimental results show that the average recognition accuracy of the system is 97.07%, and the average running time for a single ticket is 175.67ms. The practical value of the system has been tested in a commercial application, which makes a beneficial attempt for the deep learning technology in financial accounting work.
We address rotation averaging and its application to real-world 3D reconstruction. Local optimisation based approaches are the defacto choice, though they only guarantee a local optimum. Global optimizers ensure global optimality in low noise conditions, but they are inefficient and may easily deviate under the influence of outliers or elevated noise levels. We push the envelope of global rotation averaging by formulating it as a semi-definite program that can be solved efficiently by applying the Burer-Monteiro method. Both memory and time requirements are thereby largely reduced through a low-rank factorisation. Combined with a fast view graph filtering as preprocessing, and a local optimiser as post-processing, the proposed hybrid approach is robust to outliers. Compared against state-of-the-art globally optimal methods, our approach is 1 ~ 2 orders of magnitude faster while maintaining the same or better accuracy. We apply the proposed hybrid rotation averaging approach to incremental Structure from Motion (SfM) by adding the resulting global rotations as regularizers to bundle adjustment. Overall, we demonstrate high practicality of the proposed method as bad camera poses are effectively corrected and drift is reduced.
Measuring and analysing sensor data is the basic technique in vehicle dynamics development and with the advancement of embedded and data acquisition systems it is possible to analyze large data sets. In this paper a detailed method is presented for assessing and mapping isochronous trajectory patterns in Graz (Austria) by using data fusion from video, ArduinoUno and the compass sensor HDMM01. The predictive isochronous trajectory patterns are derived from the data values for a pre-defined time horizon. Both extreme driving behaviour and hazardous road geometries can be identified. It is possible to provide instant road sensor data which can be used to compare the data from a trajectory path as well as for different time instances. Results of this study show that the trajectory patterns are successful in predicting the likely evolution of a current trajectory pattern and can provide assessment on future driving situations. The obtained data from this study can be useful as reference in future city planning for energy saving driving pathways as well as vehicle design and engineering improvements based on quantitative and relevant dynamic measurements.
Positioning systems and frameworks use various techniques to determine the position of an object. Some of the existing solutions combine different sensory data at the time of positioning in order to compute more accurate positions by reducing the error introduced by the used individual positioning techniques. We present OpenHPS, a generic hybrid positioning system implemented in TypeScript, that can not only reduce the error during tracking by fusing different sensory data based on different algorithms, but also also make use of combined tracking techniques when calibrating or training the system. In addition to a detailed discussion of the architecture, features and implementation of the extensible open source OpenHPS framework, we illustrate the use of our solution in a demonstrator application fusing different positioning techniques. While OpenHPS offers a number of positioning techniques, future extensions might integrate new positioning methods or algorithms and support additional levels of abstraction including symbolic locations.
We devise a machine learning technique to solve the general problem of inferring network links that have time-delays. The goal is to do this purely from time-series data of the network nodal states. This task has applications in fields ranging from applied physics and engineering to neuroscience and biology. To achieve this, we first train a type of machine learning system known as reservoir computing to mimic the dynamics of the unknown network. We formulate and test a technique that uses the trained parameters of the reservoir system output layer to deduce an estimate of the unknown network structure. Our technique, by its nature, is non-invasive, but is motivated by the widely-used invasive network inference method whereby the responses to active perturbations applied to the network are observed and employed to infer network links (e.g., knocking down genes to infer gene regulatory networks). We test this technique on experimental and simulated data from delay-coupled opto-electronic oscillator networks. We show that the technique often yields very good results particularly if the system does not exhibit synchrony. We also find that the presence of dynamical noise can strikingly enhance the accuracy and ability of our technique, especially in networks that exhibit synchrony.
Challenges persist in nonholonomic robot navigation in dynamic environments. This paper presents a framework for such navigation based on the model of generalized velocity obstacles (GVO). The idea of velocity obstacles has been well studied and developed for obstacle avoidance since being proposed in 1998. Though it has been proved to be successful, most studies have assumed equations of motion to be linear, which limits their application to holonomic robots. In addition, more attention has been paid to the immediate reaction of robots, while advance planning has been neglected. By applying the GVO model to differential drive robots and by combining it with RRT*, we reduce the uncertainty of the robot trajectory, thus further reducing the range of concern, and save both computation time and running time. By introducing uncertainty for the dynamic obstacles with a Kalman filter, we dilute the risk of considering the obstacles as uniformly moving along a straight line and guarantee the safety. Special concern is given to path generation, including curvature check, making the generated path feasible for nonholonomic robots. We experimentally demonstrate the feasibility of the framework.
For future learning systems incremental learning is desirable, because it allows for: efficient resource usage by eliminating the need to retrain from scratch at the arrival of new data; reduced memory usage by preventing or limiting the amount of data required to be stored -- also important when privacy limitations are imposed; and learning that more closely resembles human learning. The main challenge for incremental learning is catastrophic forgetting, which refers to the precipitous drop in performance on previously learned tasks after learning a new one. Incremental learning of deep neural networks has seen explosive growth in recent years. Initial work focused on task incremental learning, where a task-ID is provided at inference time. Recently we have seen a shift towards class-incremental learning where the learner must classify at inference time between all classes seen in previous tasks without recourse to a task-ID. In this paper, we provide a complete survey of existing methods for incremental learning, and in particular we perform an extensive experimental evaluation on twelve class-incremental methods. We consider several new experimental scenarios, including a comparison of class-incremental methods on multiple large-scale datasets, investigation into small and large domain shifts, and comparison on various network architectures.