The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Deep Convolutional Neural Networks based object detection has made significant progress recent years. However, detecting small scale objects is still a challenging task. This paper addresses the problem and proposes a unified deep neural network building upon the prominent Faster R-CNN framework. This paper has two main contributions. Firstly, an Atrous Region Proposal Network (ARPN) is proposed to...
Conventional deep neural networks (DNN) for speech acoustic modeling rely on Gaussian mixture models (GMM) and hidden Markov model (HMM) to obtain binary class labels as the targets for DNN training. Subword classes in speech recognition systems correspond to context-dependent tied states or senones. The present work addresses some limitations of GMM-HMM senone alignments for DNN training. We hypothesize...
In this paper, we propose a supervised dictionary learning algorithm that aims to preserve the local geometry in both dimensions of the data. A graph-based regularization explicitly takes into account the local manifold structure of the observations. A second graph regularization gives similar treatment to the feature domain and helps in learning a more robust dictionary. Both graphs can be constructed...
Active learning aims to reduce the time and cost of developing speech recognition systems by selecting for transcription highly informative subsets from large pools of audio data. Previous evaluations at OpenKWS and IARPA BABEL have investigated data selection for low-resource languages in very constrained scenarios with 2-hour data selections given a 1-hour seed set. We expand on this to investigate...
This paper introduces a pre-training technique for learning discriminative features from electroencephalography (EEG) recordings using deep neural networks. EEG data are generally only available in small quantities, they are high-dimensional with a poor signal-to-noise ratio, and there is considerable variability between individual subjects and recording sessions. Similarity-constraint encoders as...
Compared with H.264, High Efficient Video Coding (HEVC) improves the coding efficiency by 50% at the price of significant increase in encoding time, due to Rate Distortion Optimization (RDO) on large variations of block sizes and prediction modes. In this paper, a fast intra coding algorithm is proposed to alleviate the high computational complexity of HEVC intra-frame coding. The proposed algorithm...
Face recognition systems are designed to handle well-aligned images captured under controlled situations. However real-world images present varying orientations, expressions, and illumination conditions. Traditional face recognition algorithms perform poorly on such images. In this paper we present a method for face recognition adapted to real-world conditions that can be trained using very few training...
Time-based Spiking Neural Network (SNN) has recently received increased attentions in neuromorphic computing system designs due to more bio-plausibility and better energy-efficiency. However, unleashing its potentials in realistic cognitive applications is facing significant challenges such as inefficient information representations and impractical learnings. In this work, we aim for exploring a practical...
In our earlier work, we have explored the sparse representation classification (SRC) for language recognition (LR) task. In those works, the orthogonal matching pursuit (OMP) algorithm was used for sparse coding. In place of l0-norm minimization in the OMP algorithm, one could also use ll-norm minimization based sparse coding such as the least absolute shrinkage and selection operator (LASSO). Though...
In this work, we explore the use of sparse features derived using a learned dictionary for language recognition (LR). These sparse features are referred to as s-vector and are derived by sparse coding of the commonly used low-dimensional i-vector based representation of speech utterances over the learned dictionary. The orthogonal matching pursuit (OMP), least absolute shrinkage and selection operator...
Despite being popular end-user tools, spreadsheets suffer from the vulnerability of error-proneness. In software engineering, testing has been proposed as a way to address errors. It is important therefore to know whether spreadsheet users also test, or how do they test and to what extent, especially since most spreadsheet users do not have the training, or experience, of software engineering principles...
Design Patterns for Object Oriented Systems constitute an important tool for improving software quality by providing reusable design. Many academic institutions believe in their relevance, and do courses accordingly. This paper explores practitioners' perception of the relevance their patterns knowledge has for their work. The paper also assesses how managers' perception of pattern knowledge conforms...
Convolutional Neural Networks (CNNs) have emerged as a fundamental technology for machine learning. High performance and extreme energy efficiency are critical for deployments of CNNs, especially in mobile platforms such as autonomous vehicles, cameras, and electronic personal assistants. This paper introduces the Sparse CNN (SCNN) accelerator architecture, which improves performance and energy efficiency...
Spatial separation of suspended particles based on contrast in their physical or chemical properties forms the basis of various biological assays performed on lab-on-a-chip devices. To electronically acquire this information, we have recently introduced a microfluidic sensing platform, called Microfluidic CODES, which combines the resistive pulse sensing with the code division multiple access in multiplexing...
A lot of artifiicial neural networks were proposed by scientists over the last time. Each of them can cope with the tasks of limited difficulty level, determined by their properties and capabilities. The aim of this paper is to outline difference of them and to define their positive and negative sites in different tasks of identification and control.
The Factored 3-way Restricted Boltzmann Machine has encoded the image transformation successfully. But when utilize the code to unknown image, the result was much affected by the feature of training samples. Based on the model, we separated the transformation feature out of the hidden representation and designed a new probabilistic model with gate for learning distributed representations of image...
Credit risk analysis seeks to determine whether a customer is likely to default on the financial obligation, which is a very important problem in finance. In this paper, we will present a machine learning framework to deal with this problem by formulating it as a binary classification problem. The framework consists of two parts: dictionary learning and classifier training. Firstly, we introduce a...
We propose mutually incoherent pose bases for action recognition in static image, each of which implicitly represents co-occurrence of poselets. First of all, action specific poselets are trained. To suppress the ambiguity of detection, we cluster poselet activations by the overlap of predicted torso bound of each poselet. Then pose feature of an action person can be extracted which is a vector composed...
We present Deep Sparse-coded Network (DSN), a deep architecture based on multilayer sparse coding. It has been considered difficult to learn a useful feature hierarchy by stacking sparse coding layers in a straightforward manner. The primary reason is the modeling assumption for sparse coding that takes in a dense input and yields a sparse output vector. Applying a sparse coding layer on the output...
Error Correcting Output Coding (ECOC) is a multi-class classification technique in which multiple binary classifiers are trained according to a preset code matrix such that each one learns a separate dichotomy of the classes. While ECOC is one of the best solutions for multi-class problems, one issue which makes it suboptimal is that the training of the base classifiers is done independently of the...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.