scholarly journals Optimal learning with excitatory and inhibitory synapses

2020 ◽  
Vol 16 (12) ◽  
pp. e1008536
Author(s):  
Alessandro Ingrosso

Characterizing the relation between weight structure and input/output statistics is fundamental for understanding the computational capabilities of neural circuits. In this work, I study the problem of storing associations between analog signals in the presence of correlations, using methods from statistical mechanics. I characterize the typical learning performance in terms of the power spectrum of random input and output processes. I show that optimal synaptic weight configurations reach a capacity of 0.5 for any fraction of excitatory to inhibitory weights and have a peculiar synaptic distribution with a finite fraction of silent synapses. I further provide a link between typical learning performance and principal components analysis in single cases. These results may shed light on the synaptic profile of brain circuits, such as cerebellar structures, that are thought to engage in processing time-dependent signals and performing on-line prediction.

2017 ◽  
Vol 119 (13) ◽  
pp. 1-20
Author(s):  
Marcel V. J. Veenman

Metacognitive skills refers to individual abilities for regulating and controlling learning behavior. Orientation, goal setting, planning, monitoring, and evaluation are manifestations of those skills. Given that metacognitive skills directly affect learning behavior, they are a strong predictor of learning performance. Students display a huge variation in metacognitive skillfulness, dependent on age and experience. In this article, metacognitive skills are considered to be an acquired program of self-instructions, that is, an orderly series of condition-action rules that contain conditional knowledge about when to apply which skill, and operational instructions for how to implement a particular skill. This notion has implications for effective metacognitive instruction in deficient students. Prior to instruction, on-line assessments of metacognitive skillfulness during actual task performance are indispensable for the identification of deficient students and for tailoring metacognitive instruction to the individual needs of students. Instruction should subsequently address what skill to perform when, why, and how (WWW&H), embedded within the context of a given task. Moreover, instruction should explicitly inform students about the benefits of applying metacognitive skills to make them exert the required effort. Finally, teachers may act as role model to students by including explicit metacognitive instruction in their lessons.


Groupwork ◽  
2020 ◽  
Vol 2 (1) ◽  
pp. 58-69
Author(s):  
Barry M Daste

This article attempts to shed light on some of the problems involved in developing optimum service groups for cancer patients and offers ideas concerning the design, content, leadership and membership of these groups. The article begins with a literature review of current research on issues faced by cancer patients and how these have been handled in support groups and therapy groups across the country. Following this, suggestions are offered to assist those involved in planning for these groups to deal with some of the potential difficulties encountered by many of these groups. Interest in this project grew out of the author’s personal experience with cancer and from the experience of being first a participant, and later a leader, in groups for cancer patients.Publisher’s note: We are now putting all back issues of Groupwork on line. Articles in this issue have been scanned to pdf files as viable original typesetting files no longer exist. Though they may not look it, these files are to some extent searchable. This issue was published nearly 30 years ago. We have stated author professional details as received at time of publication.


2019 ◽  
Vol 18 (01) ◽  
pp. 109-127
Author(s):  
Ting Hu ◽  
Jun Fan ◽  
Dao-Hong Xiang

In this paper, we establish the error analysis for distributed pairwise learning with multi-penalty regularization, based on a divide-and-conquer strategy. We demonstrate with [Formula: see text]-error bound that the learning performance of this distributed learning scheme is as good as that of a single machine which could process the whole data. With semi-supervised data, we can relax the restriction of the number of local machines and enlarge the range of the target function to guarantee the optimal learning rate. As a concrete example, we show that the work in this paper can apply to the distributed pairwise learning algorithm with manifold regularization.


2011 ◽  
Vol 9 (4) ◽  
pp. 41-56 ◽  
Author(s):  
Ruey-Shiang Shaw ◽  
Huan-Chao Keh ◽  
Nan-Ching Huang

Information Security Awareness, though known as a primary and important issue in the domain of Information Security, CSI computer crime and security survey showed poor security awareness training in public and private sectors. In many studies, the authors have found that the usage of knowledge maps helps the process of learning and conception building. Therefore, the authors have tried to figure out if the implementation of knowledge maps on the training materials of Information Security Awareness will improve the learning performance. Based on the e-learning materials for E-Mail Information Security Awareness, the authors designed them in the format of knowledge maps and compared the learning performance with common browsing-based materials. The results showed the knowledge map-based materials lead to higher learning performance than browse-based ones.


2011 ◽  
Vol 520 (1) ◽  
pp. 130-141 ◽  
Author(s):  
Annarita Patrizi ◽  
Laura Viltono ◽  
Elena Frola ◽  
Kirsten Harvey ◽  
Robert J. Harvey ◽  
...  

Author(s):  
Wahyuniati Hamid ◽  
La Ode Anto ◽  
Nasrul Nasrul

The study attempts to shed light on factors driving people to turn to sharia banks. The study focuses on consumer innovativeness with alternative capacity and value attractiveness as antecedents. The respondents are sharia banking consumers in Makassar. The sample size follows Malhotra 2007 formula. Respondents are reached through on-line interaction and offline contact on the spot of sharia banks. It applies the PLS tool for data analysis. It conceives that alternative seeking and innovativeness have significant effects on consumer innovativeness and desire to try the transactions with sharia banking, and consumer innovativeness has a significant effect on the desire to try the transactions with sharia banks. In this way, it explores the mediating role of consumer innovativeness in the relationship between alternative capacity and desire and that between value attractiveness and desire. Thus, the study has several novelties. It brings up new constructs such as alternative capacity, value attractiveness, and desire to try the transactions with sharia banking. The results would be that consumer innovativeness serves as a partial mediator in the relationship between alternative capacity and the desire, and a full mediator in that between value attractiveness and the desire.


1995 ◽  
Vol 6 (5) ◽  
pp. 262-270 ◽  
Author(s):  
Steven B. Greene ◽  
Gail McKoon

An interpersonal verb such as annoy or admire can be categorized according to whether its grammatical subject or grammatical object initiates the interaction described by the verb Such a verb can also be categorized according to whether a derived adjective describes its grammatical subject (e g, annoying) or its grammatical object (e g, admirable) Although there has been much speculation (e g, Brown & Fish, 1983) that these and other characteristics of these verbs shed light on basic principles of human social interaction, we argue that research to date has failed to demonstrate directly any real-time consequences of these verbs during language comprehension We present evidence that the initiating-reacting distinction predicts on-line changes in the accessibility of these verbs' arguments, but that the existence of a derived adjective does not We conclude that tasks that question subjects explicitly about language may fail to reflect the ordinary processes of language comprehension


2015 ◽  
Vol 25 (2) ◽  
pp. 361-375 ◽  
Author(s):  
David Gañán ◽  
Santi Caballé ◽  
Jordi Conesa ◽  
Fatos Xhafa

Abstract This contribution proposes software infrastructure to support new types of learning methodologies and resources based on collaborative knowledge engineering by means of an innovative application framework called the virtualized collaborative sessions framework (VCSF). The VCSF helps meet challenging collaborative knowledge engineering requirements in online learning, such as increasing group members’ learning performance during the on-line collaborative learning process. In turn, systematic application of the VCSF platform enriched with semantic knowledge engineering technologies enables e-learning developers to leverage successful collaborative learning experiences in a software reuse fashion while saving development time and effort. The framework is prototyped and successfully tested in real environments, thus showing the software reuse capability and the collaborative knowledge engineering benefits of the VCSF approach. The research reported in this paper was undertaken within the ALICE project funded through the European 7th Framework Program (FP7).


Sign in / Sign up

Export Citation Format

Share Document