Abstract-Efforts to achieve the long-standing dream of realizing scalable learning algorithms for networks of spiking neurons in silicon have been hampered by (a) the limited scalability of analog neuron circuits; (b) the enormous area overhead of learning circuits, which grows with the number of synapses; and (c) the need to implement all inter-neuron communication via off-chip address-events. In this work, a new architecture is proposed to overcome these challenges by combining innovations in computation, memory, and communication, respectively, to leverage (a) robust digital neuron circuits; (b) novel transposable SRAM arrays that share learning circuits, which grow only with the number of neurons; and (c) crossbar fan-out for efficient on-chip inter-neuron communication. Through tight integration of memory (synapses) and computation (neurons), a highly configurable chip comprising 256 neurons and 64K binary synapses with on-chip learning based on spike-timing dependent plasticity is demonstrated in 45nm SOI-CMOS. Near-threshold, event-driven operation at 0.53V is demonstrated to maximize power efficiency for real-time pattern classification, recognition, and associative memory tasks. Future scalable systems built from the foundation provided by this work will open up possibilities for ubiquitous ultra-dense, ultra-low power brain-like cognitive computers.

Location-based social networks (LBSNs) offer researchers rich data to study people's online activities and mobility patterns. One important application of such studies is to provide personalized point-of-interest (POI) recommendations to enhance user experience in LBSNs. Previous solutions directly predict users' preference on locations but fail to provide insights about users' preference transitions among locations. In this work, we propose a novel categoryaware POI recommendation model, which exploits the transition patterns of users' preference over location categories to improve location recommendation accuracy. Our approach consists of two stages: (1) preference transition (over location categories) prediction, and (2) category-aware POI recommendation. Matrix factorization is employed to predict a user's preference transitions over categories and then her preference on locations in the corresponding categories. Real data based experiments demonstrate that our approach outperforms the state-of-the-art POI recommendation models by at least 39.75% in terms of recall.

Recommendation accuracy can be improved by incorporating trust relationships derived from social networks. Most recent work on social network based recommendation is focused on minimizing the root mean square error (RMSE). Social network based top-k recommendation, which recommends to a user a small number of items at a time, is not well studied. In this paper, we conduct a comprehensive study on improving the accuracy of top-k recommendation using social networks. We first show that the existing social-trust enhanced Matrix Factorization (MF) models can be tailored for top-k recommendation by including observed and missing ratings in their training objective functions. We also propose a Nearest Neighbor (NN) based top-k recommendation method that combines users' neighborhoods in the trust network with their neighborhoods in the latent feature space. Experimental results on two publicly available datasets show that social networks can significantly improve the top-k hit ratio, especially for cold start users. Surprisingly, we also found that the technical approach for combining feedback data (e.g. ratings) with social network information that works best for minimizing RMSE works poorly for maximizing the hit ratio, and vice versa.

Mobile video broadcasting service, or mobile TV, is expected to become a popular application for 3G wireless network operators. Most existing solutions for video Broadcast Multicast Services (BCMCS) in 3G networks employ a single transmission rate to cover all viewers. The system-wide video quality of the cell is therefore throttled by a few viewers close to the boundary, and is far from reaching the social-optimum allowed by the radio resources available at the base station. In this paper, we propose a novel scalable video broadcast/multicast solution, SV-BCMCS, that efficiently integrates scalable video coding, 3G broadcast and ad-hoc forwarding to balance the system-wide and worst-case video quality of all viewers at 3G cell. In our solution, video is encoded into multiple layers. The base station broadcasts different layers at different rates to cover viewers at different ranges. All viewers are guaranteed to receive the base layer, and viewers closer to the base station can receive more enhancement layers. Using ad-hoc connections, viewers far away from the base station can obtain from their neighbors closer to the base station the enhancement layers that they cannot receive directly from the base station. We study the optimal resource allocation problem in SV-BCMCS and develop practical helper finding and relay routing algorithms. Through analysis and extensive OPNET simulations, we demonstrated that SV-BCMCS can significantly improve the system-wide video quality at the price of slight quality degradation of a few viewers close to the boundary.

The present paper aims at getting the porous effect parameter G of a thin permeable wall. The reflection and transmission coefficients of a thin vertical porous wall with different porous shapes and porosities are obtained by a physical model experiment. The reflection coefficient and total horizontal wave load on a perforated caisson are also obtained by another experiment. Analytical solutions of wave interaction with these two vertical porous breakwaters are introduced. The reflection and transmission coefficients between predictions and experimental results of authors' and four other researchers' are compared. By comparisons, the estimate methods of the linearized resistance coefficient f and the inertial effect coefficient s of the permeable plate are recommended. Following the formula of Yu [1995], the values of G are calculated by f and s. A complete estimate method of porous effect parameter G is presented, which consists of the formula of Yu [1995] and the present experimental formula. The estimate method of G is validated by comparing * Corresponding author. 309 Coast. Eng. J. 2006.48:309-336. Downloaded from www.worldscientific.com by UNIVERSITY OF CALIFORNIA @ SAN DIEGO on 01/03/15. For personal use only. 310 Y. Li, Y. Liu & B. Teng the predicted values of the total horizontal wave loads on the perforated caisson with the experimental results.

No abstract

No abstract

No abstract