Helsinki University of Technology

About: Helsinki University of Technology is a based out in . It is known for research contribution in the topics: Artificial neural network & Finite element method. The organization has 8962 authors who have published 20136 publications receiving 723787 citations. The organization is also known as: TKK & Teknillinen korkeakoulu.

Comparing subspace clusterings

Abstract: We present the first framework for comparing subspace clusterings. We propose several distance measures for subspace clusterings, including generalizations of well-known distance measures for ordinary clusterings. We describe a set of important properties for any measure for comparing subspace clusterings and give a systematic comparison of our proposed measures in terms of these properties. We validate the usefulness of our subspace clustering distance measures by comparing clusterings produced by the algorithms FastDOC, HARP, PROCLUS, ORCLUS, and SSPC. We show that our distance measures can be also used to compare partial clusterings, overlapping clusterings, and patterns in binary data matrices.

Efficient digital implementation of the sigmoid function for reprogrammable logic

Abstract: Special attention must be paid to an efficient approximation of the sigmoid function in implementing FPGA-based reprogrammable hardware-based artificial neural networks. Four previously published piecewise linear and one piecewise second-order approximation of the sigmoid function are compared with SIG-sigmoid, a purely combinational approximation. The approximations are compared in terms of speed, required area resources and accuracy measured by average and maximum error. It is concluded that the best performance is achieved by SIG-sigmoid.

Synthesis and stability of monolayer-protected Au38 clusters.

Abstract: A synthesis strategy to obtain monodisperse hexanethiolate-protected Au38 clusters based on their resistance to etching upon exposure to a hyperexcess of thiol is reported. The reduction time in the standard Brust−Schiffrin two-phase synthesis was optimized such that Au38 were the only clusters that were fully passivated by the thiol monolayer which leaves larger particles vulnerable to etching by excess thiol. The isolated Au38 was characterized by mass spectrometry, thermogravimetric analysis, optical spectroscopy, and electrochemical techniques giving Au38(SC6)22 as the molecular formula for the cluster. These ultrasmall Au clusters behave analogously to molecules with a wide energy gap between occupied (HOMO) and unoccupied levels (LUMO) and undergo single-electron charging at room temperature in electrochemical experiments. Electrochemistry provides an elegant means to study the electronic structure and the chemical stability of the clusters at different charge states. We used cyclic voltammetry and ...

On the feasibility of full-duplex relaying in the presence of loop interference

Abstract: We consider a two-hop infrastructure-based relay link with three transmit power allocation methods. Both amplify-and-forward and decode-and-forward protocols are studied. Fundamentally, a relay can operate either in a full-duplex or in a half-duplex mode. The performance in full-duplex relaying is degraded due to loop interference from the relay output to the relay input. On the other hand, the half-duplex mode eliminates the loop interference, but this reduces the end-to-end rate. In this paper, we study the feasibility of the full-duplex mode in terms of end-to-end capacity. In particular, we evaluate break-even loop interference levels, below which the full-duplex mode outperforms the half-duplex mode. The analysis shows that the full-duplex mode is feasible in the presence of reasonably high loop interference power and can offer higher capacity than the half-duplex mode.