This paper presents two novel formulations for learning shared feature representations across multiple tasks. The idea is to pose the problem as that of learning a shared kernel, which is constructed from a given set of base kernels, leading to improved generalization in all the tasks. The first formulation employs a (l1, lp), p ≥ 2 mixed norm regularizer promoting sparse combinations of the ba...

Transconductors are commonly used as active elements in high-frequency (HF) filters, amplifiers, mixers, and oscillators. This paper reviews transconductor design by focusing on the V-I kernel that determines the key transconductor properties. Based on bandwidth considerations, simple V-I kernels with few or no internal nodes are preferred. In a systematic way, virtually all simple kernels publ...

Making use of the Volterra approach, black box modeling is applied to a large scale analog circuitry for an ADSL (asymmetric digital subscriber line) central office application. Reducing the number of free parameters through special assumptions on the Volterra kernels, one ends up with the Hammerstein model. Taking the first few taps of the Volterra kernels and approximating the last taps throu...

Production of fluid from or injection into a coal seam leads to variations permeability the seam, resulting adsorption in matrix particular. One subtle effect is that, if pressure cleat increases, after an immediate opening cleat, one expects aperture (and hence permeability) decrease for transient time, as consequence diffusion matrix. In this work, we aim model such by proposing constitutive ...

In the present article, we extend our study of Kantorovich type exponential sampling operators introduced in [4]. We derive Voronovskaya theorem and its quantitative estimates for these terms an appropriate K-functional. Further, improve order approximation by using convex linear combinations operators. Finally, provide few examples kernels along with graphical representations.

Compared with word-level and sentence-level convolutional neural networks (ConvNets), the character-level ConvNets has a better applicability for misspellings and typos input. Due to this, recent researches for text classification mainly focus on character-level ConvNets. However, while the majority of these researches employ English corpus for the character-level text classification, few resea...

Sparse Matrix is a matrix consisting of very few non-zero entries. Large sparse matrices are often used in engineering and scientific operations. Especially sparse-matrix vector multiplication is an important operation for solving linear system and partial differential equations. However, there is a possibility that even though the matrix is partitioned and stored appropriately, the performance...

In this paper, we propose a discriminative approach for retrieval of video shots characterized by a sequential structure. The task of retrieving shots similar in content to a few positive example shots is more close to a binary classification problem. Hence, this task can be solved by a discriminative learning approach. For a content-based retrieval task the twin characteristics of rare positiv...

Tests were conducted to determine the effects of drying conditions on the rate of fissure formation after drying and the resultant mechanical strength distributions of individual rice kernels. Long-grain varieties Cypress, Drew, and Wells at 21% harvest moisture content (HMC) were dried at 40°C, 45°C, 50°C, 55°C, and 60°C and constant 17% relative humidity (RH) to 12% MC. Immediately after dryi...

Classification of biological sequences with kernel methods Kernels and kernel methods Kernels for biological sequences Summary Outline 1 Kernels and kernel methods Kernels Kernel Methods 2 Kernels for biological sequences Motivations Feature space approach Using generative models Derive from a similarity measure Application: remote homology detection Jean-Philippe Vert Classification of biologi...