Technology learning for renewable energy: Implications for South Africa's long-term mitigation scenarios

Technology learning can make a significant difference to renewable energy as a mitigation option in South Africa's electricity sector. This article considers scenarios implemented in a Markal energy model used for mitigation analysis. It outlines the empirical evidence that unit costs of renewable energy technologies decline, considers the theoretical background and how this can be implemented in modeling. Two scenarios are modelled, assuming 27% and 50% of renewable electricity by 2050, respectively. The results show a dramatic shift in the mitigation costs. In the less ambitious scenario, instead of imposing a cost of Rand 52/t CO₂-eq (at 10% discount rate), reduced costs due to technology learning turn renewables into negative cost option. Our results show that technology learning flips the costs, saving R143. At higher penetration rate, the incremental costs added beyond the base case decline from R92 per ton to R3. Including assumptions about technology learning turns renewable from a higher-cost mitigation option to one close to zero. We conclude that a future world in which global investment in renewables drives down unit costs makes it a much more cost-effective and sustainable mitigation option in South Africa.     

Projection Learning

A method of combining learning algorithms is described that preserves attribute-efficiency. It yields learning algorithms that require a number of examples that is polynomial in the number of relevant variables and logarithmic in the number of irrelevant ones. The algorithms are simple to implement and realizable on networks with a number of nodes linear in the total number of variables. They include generalizations of Littlestone's Winnow algorithm, and are, therefore, good candidates for experimentation on domains having very large numbers of attributes but where nonlinear hypotheses are sought.     

A recurrent stochastic binary network

Stochastic neural networks are usually built by introducing random fluctuations into the network. A natural method is to use stochastic connections rather than stochastic activation functions. We propose a new model in which each neuron has very simple functionality but all the connections are stochastic. It is shown that the stationary distribution of the network uniquely exists and it is approxi-mately a Boltzmann-Gibbs distribution. The relationship between the model and the Markov random field is discussed. New techniques to implement simulated annealing and Boltzmann learning are pro-posed. Simulation results on the graph bisection problem and image recognition show that the network is powerful enough to solve real world problems.     

Learning to ignore the mask in texture segmentation tasks.

Although traditionally texture segmentation has been regarded as an automatic, preattentive process, participants confronted with texture segmentation in experimental settings (i.e., with brief presentation time and subsequent masking) are initially unable to perform the task. According to perceptual learning concepts, participants must learn to fine-tune their sensory channels before perception improves under restricted viewing conditions. The present article proposes an alternative perspective that emphasizes the role of the mask. Four experiments showed that the amount of observed learning depends on the structural and temporal homogeneity or heterogeneity of the mask. The authors suggest that learning consists of separating the task-relevant signal stemming from the texture from the task-irrelevant signal of the mask and of ignoring the mask. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Collaborative Technology in the Classroom: A Review of the GSS Research and a Research Framework

Over the past several years, a number of research studies have investigated the application of group support system (GSS) technology to the classroom. The purpose of this paper is twofold: (1) to summarize and assess what has been learned from this emerging body of research, and (2) to draw on the GSS and cooperative learning literatures to construct a framework to guide future studies. The review of research studies reported in this paper focuses on comparative studies investigating the synchronous GSS classroom environment. While research findings regarding the impacts of GSS on learning performance are somewhat mixed, early indications are encouraging and suggest that GSS may have a positive impact under certain situations. A framework based on the GSS and cooperative learning research is introduced that identifies the contextual, group process, and outcome factors that researchers may wish to consider when designing and interpreting future research studies in this area.     

Robust Learning with Missing Data

This paper introduces a new method, called the robust Bayesian estimator (RBE), to learn conditional probability distributions from incomplete data sets. The intuition behind the RBE is that, when no information about the pattern of missing data is available, an incomplete database constrains the set of all possible estimates and this paper provides a characterization of these constraints. An experimental comparison with two popular methods to estimate conditional probability distributions from incomplete data—Gibbs sampling and the EM algorithm—shows a gain in robustness. An application of the RBE to quantify a naive Bayesian classifier from an incomplete data set illustrates its practical relevance.     

Review of Learning Disabilities and Psychic Conflict: A Psychoanalytic Casebook.

Review of book: Learning Disabilities and Psychic Conflict: A Psychoanalytic Casebook by Arden Aibel Rothstein and Jules Glenn, Madison, NJ: International Universities Press, 1999, 504 pp. Reviewed by Lucy M. Zabarenko. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Adaptive Disk Spindown via Optimal Rent-to-Buy in Probabilistic Environments

In the single rent-to-buy decision problem, without a priori knowledge of the amount of time a resource will be used we need to decide when to buy the resource, given that we can rent the resource for $1 per unit time or buy it once and for all for $c . In this paper we study algorithms that make a sequence of single rent-to-buy decisions, using the assumption that the resource use times are independently drawn from an unknown probability distribution. Our study of this rent-to-buy problem is motivated by important systems applications, specifically, problems arising from deciding when to spindown disks to conserve energy in mobile computers [4], [13], [15], thread blocking decisions during lock acquisition in multiprocessor applications [7], and virtual circuit holding times in IP-over-ATM networks [11], [19]. We develop a provably optimal and computationally efficient algorithm for the rent-to-buy problem. Our algorithm uses time and space, and its expected cost for the t th resource use converges to optimal as , for any bounded probability distribution on the resource use times. Alternatively, using O(1) time and space, the algorithm almost converges to optimal. We describe the experimental results for the application of our algorithm to one of the motivating systems problems: the question of when to spindown a disk to save power in a mobile computer. Simulations using disk access traces obtained from an HP workstation environment suggest that our algorithm yields significantly improved power/ response time performance over the nonadaptive 2-competitive algorithm which is optimal in the worst-case competitive analysis model. Received October 22, 1996; revised September 25, 1997.     

广义计算系统研究

该文简要介绍运用广义计算技术而建立起来的一类人工智能系统──广义计算系统，包括广义计算系统的数学描述，广义计算系统的实现问题和学习问题。     

区块链赋能物联网中联合资源分配与控制的智能计算迁移研究

大数据场景下,远程云服务器通常被部署用于数据处理与价值挖掘,但在面对时延敏感型或需要动态频繁交互的业务时,该种处理模式显得力不从心.作为对云计算模式的补充,雾计算因其可有效降低任务处理时延、能耗与带宽消耗而备受关注;同时,面向雾计算的计算迁移机制因其能有效缓解节点的处理负担并改善用户体验而成为领域研究焦点.在雾计算模式...