We propose a recurrent extension of the Ladder networks [22] whose structure is motivated by the inference required in hierarchical latent variable models. We demonstrate that the recurrent Ladder is able to handle a wide variety of complex learning tasks that benefit from iterative inference and temporal modeling. The architecture shows close-to-optimal results on temporal modeling of video da...

It has been conjectured that the group algebra Q[G] of a discrete group G over the rational numbers Q is semi-simple. In this paper we consider irreducible representations of this algebra satisfying certain finiteness conditions and see just how much information these yield towards solving the above problem. We show in fact for a finitely generated group G, Q[G] has "enough" of these representa...

Image blurring artifact is the main challenge to any spatial, denoising filters. This artifact is contributed by the heterogeneous intensities within the given neighborhood or window of fixed size. Selection of most similar intensities (G-Neighbors) helps to adapt the window shape which is of edge-aware nature and subsequently reduce this blurring artifact. The paper presents a memristive circu...

The semi-affine Coxeter-Dynkin graph is introduced, generalizing both the affine and the finite types. Semi-affine graphs. It is profitable to treat the so-called Coxeter-Dynkin diagrams as graphs. A classification of finite graphs with an adjacency matrix having 2 as the largest eigenvalue is made in a paper of John Smith [JHS]. It is in a combinatorial context and no reference is made to Coxe...

In this paper, we introduce a way of constructing a rough set via grill ordered topological spaces. Increasing and decreasing sets are defined based on grill and comparisons between current approximations and previous approximations by Shafei and Kandil are carried out. Also it is shown that the chances of getting better approximation by our method of approximations are greater than any of the ...

For every segmentation task, prior knowledge about the object that shall be segmented has to be incorporated. This is typically performed either automatically by using labeled data to train the used algorithm, or by manual adaptation of the algorithm to the specific application. For the segmentation of 3D data, the generation of training sets is very tedious and time consuming, since in most ca...

Negotiation has been extensively discussed in electronic commerce for decades. Recent growing interest in importing machine learning algorithm in electronic commerce has given increased importance to automated negotiation. A Tri-Training based algorithm was proposed to learn opponent’s negotiation preference. The process of negotiation was viewed as a proposal’s sequence which can be mapped int...

In this study we address the problem of training a neuralnetwork for language identification using both labeled and unlabeled speech samples in the form of i-vectors. We propose a neural network architecture that can also handle out-of-set languages. We utilize a modified version of the recently proposed Ladder Network semisupervised training procedure that optimizes the reconstruction costs of...

Let G = (V, E) be a graph with V = {1, 2, . . . , n}, in which we allow parallel edges but no loops, and let S+(G) be the set of all positive semi-definite n × n matrices A = [ai,j ] with ai,j = 0 if i = j and i and j are non-adjacent, ai,j = 0 if i = j and i and j are connected by exactly one edge, and ai,j ∈ R if i = j or i and j are connected by parallel edges. The maximum positive semi-defi...

We consider a class of evolutionary variational inequalities arising in quasistatic frictional contact problems for linear elastic materials. We indicate sufficient conditions in order to have the existence, the uniqueness and the Lipschitz continuous dependence of the solution with respect to the data, respectively. The existence of the solution is obtained using a time-discretization method, ...