A multi-objective adaptive immune algorithm for multi-application NoC mapping

Current SoC design trends are characterized by the integration of larger amount of IPs targeting a wide range of application fields. Such multi-application systems are constrained by a set of requirements. In such scenario network-on-chips (NoC) are becoming more important as the on-chip communication structure. Designing an optimal NoC for satisfying the requirements of each individual application requires the specification of a large set of configuration parameters leading to a wide solution space. It has been shown that IP mapping is one of the most critical parameters in NoC design, strongly influencing the SoC performance. IP mapping has been solved for single application systems using single and multi-objective optimization algorithms. In this paper we propose the use of a multi-objective adaptive immune algorithm (M(2)AIA), an evolutionary approach to solve the multi-application NoC mapping problem. Latency and power consumption were adopted as the target multi-objective functions. To compare the efficiency of our approach, our results are compared with those of the genetic and branch and bound multi-objective mapping algorithms. We tested 11 well-known benchmarks, including random and real applications, and combines up to 8 applications at the same SoC. The experimental results showed that the M(2)AIA decreases in average the power consumption and the latency 27.3 and 42.1 % compared to the branch and bound approach and 29.3 and 36.1 % over the genetic approach.

Detection of gait perturbations based on proprioceptive information. Application to Limit Cycle Walkers

Walking on irregular surfaces and in the presence of unexpected events is a challenging problem for bipedal machines. Up to date, their ability to cope with gait disturbances is far less successful than humans': Neither trajectory controlled robots, nor dynamic walking machines (Limit CycleWalkers) are able to handle them satisfactorily. On the contrary, humans reject gait perturbations naturally and efficiently relying on their sensory organs that, if needed, elicit a recovery action. A similar approach may be envisioned for bipedal robots and exoskeletons: An algorithm continuously observes the state of the walker and, if an unexpected event happens, triggers an adequate reaction. This paper presents a monitoring algorithm that provides immediate detection of any type of perturbation based solely on a phase representation of the normal walking of the robot. The proposed method was evaluated in a Limit Cycle Walker prototype that suffered push and trip perturbations at different moments of the gait cycle, providing 100% successful detections for the current experimental apparatus and adequately tuned parameters, with no false positives when the robot is walking unperturbed.

Stochastic competitive learning in complex networks

Competitive learning is an important machine learning approach which is widely employed in artificial neural networks. In this paper, we present a rigorous definition of a new type of competitive learning scheme realized on large-scale networks. The model consists of several particles walking within the network and competing with each other to occupy as many nodes as possible, while attempting to reject intruder particles. The particle's walking rule is composed of a stochastic combination of random and preferential movements. The model has been applied to solve community detection and data clustering problems. Computer simulations reveal that the proposed technique presents high precision of community and cluster detections, as well as low computational complexity. Moreover, we have developed an efficient method for estimating the most likely number of clusters by using an evaluator index that monitors the information generated by the competition process itself. We hope this paper will provide an alternative way to the study of competitive learning.

A regional multirnodulus algorithm for blind equalization of QAM signals: Introduction and steady-state analysis

It is well known that constant-modulus-based algorithms present a large mean-square error for high-order quadrature amplitude modulation (QAM) signals, which may damage the switching to decision-directed-based algorithms. In this paper, we introduce a regional multimodulus algorithm for blind equalization of QAM signals that performs similar to the supervised normalized least-mean-squares (NLMS) algorithm, independently of the QAM order. We find a theoretical relation between the coefficient vector of the proposed algorithm and the Wiener solution and also provide theoretical models for the steady-state excess mean-square error in a nonstationary environment. The proposed algorithm in conjunction with strategies to speed up its convergence and to avoid divergence can bypass the switching mechanism between the blind mode and the decision-directed mode. (c) 2012 Elsevier B.V. All rights reserved.

The importance of landscape structure for seed dispersal in rain forest fragments

Questions What are the main features of the seed rain in a fragmented Atlantic forest landscape? Can seed rain species attributes (life form, dispersal mode, successional status) relate to the spatial arrangement (size and number of fragments, edge density and presence of corridor) of forest fragments in the landscape? How does the rain forest landscape structure affect the seed rain? Location Atlantic rainforest, Sao Paulo State, Southeastern Brazil. Methods Seed rain samples were collected monthly throughout 1yr, counted, identified and classified according to species dispersal mode, successional status and life form. Seed rain composition was compared with woody species near the seed traps. Relationships between seed rain composition and landscape spatial arrangement (fragment area, presence of corridor, number of fragments in the surroundings, proximity of fragments, and edge density) were tested using canonical correspondence analysis (CCA). Results We collected 20142 seeds belonging to 115 taxa, most of them early successional and anemochorous trees. In general, the seed rain had a species composition distinct from that of the nearby forest tree community. Small isolated fragments contained more seeds, mainly of anemochorous, epiphytic and early-successional species; large fragments showed higher association with zoochorous and late-successional species compared to small fragments. The CCA significantly distinguished the species dispersal mode according to fragment size and isolation, anemochorous species being associated to small and isolated fragments, and zoochorous species to larger areas and fragment aggregation. Nevertheless, a gradient driven by proximity (PROX) and edge density (ED) segregated lianas (in the positive extremity), early successional and epiphyte species (in the negative end); large fragments were positively associated to PROX and ED. Conclusions The results highlight the importance of the size and spatial arrangement of forest patches to promote habitat connectivity and improve the flux of animal-dispersed seeds. Landscape structure controls seed fluxes and affects plant dispersal capacity, potentially influencing the composition and structure of forest fragments. The seed rain composition may be used to assess the effects of landscape spatial structure on plant assemblages, and provide relevant information for biodiversity conservation.

MR-Radix: a multi-relational data mining algorithm

Abstract Background Once multi-relational approach has emerged as an alternative for analyzing structured data such as relational databases, since they allow applying data mining in multiple tables directly, thus avoiding expensive joining operations and semantic losses, this work proposes an algorithm with multi-relational approach. Methods Aiming to compare traditional approach performance and multi-relational for mining association rules, this paper discusses an empirical study between PatriciaMine - an traditional algorithm - and its corresponding multi-relational proposed, MR-Radix. Results This work showed advantages of the multi-relational approach in performance over several tables, which avoids the high cost for joining operations from multiple tables and semantic losses. The performance provided by the algorithm MR-Radix shows faster than PatriciaMine, despite handling complex multi-relational patterns. The utilized memory indicates a more conservative growth curve for MR-Radix than PatriciaMine, which shows the increase in demand of frequent items in MR-Radix does not result in a significant growth of utilized memory like in PatriciaMine. Conclusion The comparative study between PatriciaMine and MR-Radix confirmed efficacy of the multi-relational approach in data mining process both in terms of execution time and in relation to memory usage. Besides that, the multi-relational proposed algorithm, unlike other algorithms of this approach, is efficient for use in large relational databases.

Frequency and precision of feedback and the adaptive process of learning a dual motor task

This work investigated the effects of frequency and precision of feedback on the learning of a dual-motor task. One hundred and twenty adults were randomly assigned to six groups of different knowledge of results (KR), frequency (100%, 66% or 33%) and precision (specific or general) levels. In the stabilization phase, participants performed the dual task (combination of linear positioning and manual force control) with the provision of KR. Ten non-KR adaptation trials were performed for the same task, but with the introduction of an electromagnetic opposite traction force. The analysis showed a significant main effect for frequency of KR. The participants who received KR in 66% of the stabilization trials showed superior adaptation performance than those who received 100% or 33%. This finding reinforces that there is an optimal level of information, neither too high nor too low, for motor learning to be effective.

Cementum, apical morphology and hypercementosis: a probable adaptive response of the periodontal support tissues and potential orthodontic implications

Information about orthodontic movement of teeth with hypercementosis is scarce. As cementum deposition continues to occur, cementum is expected to change the shape of the root and apex over time, but this has not yet been demonstrated. Nor has it ever been established whether it increases or decreases the prevalence of root resorption during orthodontic treatment. The unique biological function of the interconnected network of cementocytes may play a role in orthodontic movement and its associated root resorptions, but no research has ever been conducted on the topic. Unlike cementum thickness and hypercementosis, root and apex shape has not yet been related to patient age. A study of the precise difference between increased cementum thickness and hypercementosis is warranted. Hypercementosis refers to excessive cementum formation above and beyond the extent necessary to fulfill its normal functions, resulting in abnormal thickening with macroscopic changes in the tooth root, which may require the delivery of forces that are different from conventional mechanics in their intensity, direction and distribution. What are the unique features and specificities involved in moving teeth that present with hypercementosis? Bodily movements would be expected to occur, since inclination might prove difficult to achieve, but would the root resorption index be higher or lower?