Dynamic Adaptive Neural Network Array (DANNA) is a neuromorphic hardware implementation. It differs from most other neuromorphic projects in that it allows for programmability of structure, and it is trained or design...
详细信息
Variable or value elimination in a constraint satisfaction problem (CSP) can be used in preprocessing or during search to reduce search space size. A variable elimination rule (value elimination rule) allows the polyn...
详细信息
Gene expression data are widely used in classification tasks for medical diagnosis. data scaling is recommended and helpful for learning the classification models. In this study, we propose a data scaling algorithm to...
详细信息
Air pollution attracts extensive attention globally due to its critical impacts on human life. Monitoring systems providing real-time micro-level pollution information have been developed to provide authorities with d...
详细信息
Air pollution attracts extensive attention globally due to its critical impacts on human life. Monitoring systems providing real-time micro-level pollution information have been developed to provide authorities with data to mitigate these impacts. However, current systems are usually application-specific with fixed hardware and software configurations. They are inconvenient in maintenance, infeasible in reconfiguration, and unexpandable in sensing capabilities. This paper proposes a novel Modular Sensor System (MSS), which aims at tackling these issues by adopting the proposed Universal Sensor Interface (USI) and modular design in a sensor node. A compact MSS senor node with expandable plug-and-play sensor modules and multiple Wireless Sensor Networks (WSNs) compatibility is implemented and evaluated. Results indicate that MSS sensor node can be deployed in different scenarios while dynamically adapting to reconfigurations and monitoring air pollution at low concentration levels with high energy efficiency. We anticipate that MSS is able to relax the efforts on system maintenance, adaptation, and evolution in real-life large-scale deployment situation.
With this poster, we will present the Linked SPARQL Query (LSQ) dataset, which describes SPARQL queries taken from the logs of public endpoints. We introduce the initial four query logs that we have taken and the extr...
详细信息
With this poster, we will present the Linked SPARQL Query (LSQ) dataset, which describes SPARQL queries taken from the logs of public endpoints. We introduce the initial four query logs that we have taken and the extraction process applied: the types of meta-data captured, how the data are modelled, what vocabularies we use, etc. The LSQ dataset currently contains 73 million triples describing 5.7 million query executions and is publicly available as Linked data and through a SPARQL endpoint. We believe that by providing insights on how SPARQL is used in practice, the LSQ dataset could benefit areas of SPARQL research, including caching, benchmarking, usability, optimisations, etc.
Gene expression data are widely used in classification tasks for medical diagnosis. data scaling is recommended and helpful for learning the classification models. In this study, we propose a data scaling algorithm to...
详细信息
ISBN:
(纸本)9781467379847
Gene expression data are widely used in classification tasks for medical diagnosis. data scaling is recommended and helpful for learning the classification models. In this study, we propose a data scaling algorithm to transform the data uniformly to an appropriate interval by learning a generalized logistic function to fit the empirical cumulative density function of the data. The proposed algorithm is robust to outliers, and experimental results show that models learned using data scaled by the proposed algorithm generally outperform the ones using min-max mapping and z-score which are currently the most commonly used data scaling algorithms.
The Alternating Direction Method of Multipliers (ADMM) has been studied for years. Traditional ADMM algorithms need to compute, at each iteration, an (empirical) expected loss function on all training examples, result...
详细信息
ISBN:
(纸本)9781510810587
The Alternating Direction Method of Multipliers (ADMM) has been studied for years. Traditional ADMM algorithms need to compute, at each iteration, an (empirical) expected loss function on all training examples, resulting in a computational complexity proportional to the number of training examples. To reduce the complexity, stochastic ADMM algorithms were proposed to replace the expected loss function with a random loss function associated with one uniformly drawn example plus a Bregman divergence ter-m. The Bregman divergence, however, is derived from a simple 2nd-order proximal function, i.e., the half squared norm, which could be a subopti-mal choice. In this paper, we present a new family of stochastic ADMM algorithms with optimal 2nd-order proximal functions, which produce a new family of adaptive stochastic ADMM methods. We theoretically prove that the regret bounds are as good as the bounds which could be achieved by the best proximal function that can be chosen in hindsight. Encouraging empirical results on a variety of real-world datasets confirm the effectiveness and efficiency of the proposed algorithms.
Latent Factor models, which transform both users and items into the same latent feature space, are one of the most successful and ubiquitous models in recommender systems. Most existing models in this paradigm define ...
详细信息
A crucial part of the total electricity demand is energy consumption in the residential sector. In parallel to optimizing energy consumption within houses, user comfort is still an essential success criterion for auto...
详细信息
A crucial part of the total electricity demand is energy consumption in the residential sector. In parallel to optimizing energy consumption within houses, user comfort is still an essential success criterion for automated solutions used within the house. Choosing the most comfortable appliance schedule is often a challenging task for the members of the house. To bring focus on this challenge, residential customer involvement is enhanced by a trend towards automation of appliances. This trend is reflected by pilot projects such as Linear which uses automated smart appliances at the demand side to attain more flexibility in the electricity system. Moreover, industrial interest from the Telecom, energy and household appliance sector to promote smart schedules for appliances is growing. To meet this trend, this paper describes new ways to model and reason with the user preferences when scheduling appliances in a household under dynamic pricing schemes given different user preferences. These methods have been proven to be efficient in eliciting and computing the user preferences to increase the user comfort in the house.
暂无评论