69,685 Hits in 6.0 sec

Maximizing Learning Progress: An Internal Reward System for Development [chapter]

Frédéric Kaplan, Pierre-Yves Oudeyer
2004 Lecture Notes in Computer Science  
This chapter presents a generic internal reward system that drives an agent to increase the complexity of its behavior. This reward system does not reinforce a predefined task.  ...  Its purpose is to drive the agent to progress in learning given its embodiment and the environment in which it is placed.  ...  Acknowledgements The authors would like to thank Luc Steels and the members of the developmental robotics team (Verena Hafner, Claire d'Este and Andrew Whyte) for precious comments on this work.  ... 
doi:10.1007/978-3-540-27833-7_19 fatcat:vvahfutbu5bpjhg33rqzsdmopa

Maximizing Entrepreneurship learning/training benefits

2017 Journal of finance & corporate governance  
As an educator who strives for highest quality university teaching, I do my best to create a complete immersion of my students in learning from theory, best practices, real examples and a great variety  ...  A major policy implication is the need for great improvements in courses content and teaching methods.  ...  At certain times, benchmarking and evaluation of progress could be useful to maximize or optimize learning and training of our students and youth.  ... 
doi:10.54960/jfcg.v1i1.1 fatcat:ixfozivafng7dhyogu4ndlea3a

Maximizing Learning for Students with Special needs

Deborah Edelman Watkins
2005 Kappa Delta Pi Record  
A Maximizing Learning for Students with Special Needs by Deborah Edelman Watkins that an educator must, first and foremost, develop a relationship.  ...  Such research underscores the powerful role the classroom environment can play in the development of all children's expectations for learning.  ... 
doi:10.1080/00228958.2005.10532062 fatcat:yd5dpbl5sjhorby4fqjtohtwze

Hindsight Expectation Maximization for Goal-conditioned Reinforcement Learning [article]

Yunhao Tang, Alp Kucukelbir
2021 arXiv   pre-print
We propose a graphical model framework for goal-conditioned RL, with an EM algorithm that operates on the lower bound of the RL objective.  ...  The E-step provides a natural interpretation of how 'learning in hindsight' techniques, such as HER, to handle extremely sparse goal-conditioned rewards.  ...  In International conference on machine learning, pages 1-9. PMLR, 2013.  ... 
arXiv:2006.07549v2 fatcat:p5bpmz7ktjdhnpuwjr3vmig7la

Model-Independent Online Learning for Influence Maximization [article]

Sharan Vaswani, Branislav Kveton, Zheng Wen, Mohammad Ghavamzadeh, Laks Lakshmanan, Mark Schmidt
2018 arXiv   pre-print
For this, we propose a pairwise-influence semi-bandit feedback model and develop a LinUCB-based bandit algorithm.  ...  We also consider the case of a new marketer looking to exploit an existing social network, while simultaneously learning the factors governing information propagation.  ...  Influence maximization in continuous time dif- fusion networks. In 29th International Conference on Machine Learning (ICML 2012), pp. 1-8. International Machine Learning Society, 2012.  ... 
arXiv:1703.00557v2 fatcat:grvjpn2e3bathmtqrkmvr3fdya

Speed Learning: Maximizing Student Learning and Engagement in a Limited Amount of Time

Arshia A. Khan, Janna Madden
2016 International Journal of Modern Education and Computer Science  
After the employment of a series of this active learning technique a survey of the students revealed an increase in student learning.  ...  The active learning method employed in this study is grounded in classic pedagogies that have been developed based on various psychological theories of learning, motivation and engagement.  ...  Khan, Janna Madden,"Speed Learning: Maximizing Student Learning and Engagement in a Limited Amount of Time", International Journal of Modern Education and Computer Science(IJMECS), Vol.8, No.7, pp.22-30  ... 
doi:10.5815/ijmecs.2016.07.03 fatcat:gfcc4ddefjc6han2wzra6umi2q

Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interference [article]

Matthew Riemer, Ignacio Cases, Robert Ajemian, Miao Liu, Irina Rish, Yuhai Tu, Gerald Tesauro
2019 arXiv   pre-print
proposed baselines for continual learning.  ...  Lack of performance when it comes to continual learning over non-stationary distributions of data remains a major challenge in scaling neural network learning to more human realistic settings.  ...  Additionally, we would like to thank Arslan Chaudhry and Marc'Aurelio Ranzato for their helpful comments and discussions. We also thank the three anonymous reviewers for their valuable suggestions.  ... 
arXiv:1810.11910v3 fatcat:jz53fudg7zfxvnprthdl5i6se4

Learning A La Carte: A Theory-Based Tool For Maximizing Student Engagement

Jeremy Sibold
2016 Journal of College Teaching & Learning (TLC)  
It is the purpose of this article to examine relevant SDT research, and utilize relatable trans-disciplinary findings in support of the discussion of a novel course development technique that maximizes  ...  Imagine the difference in the class experience for students and instructors alike where there is a system in place that allows all stakeholders to have control and autonomy in the course progression.  ...  rewarding, interesting, and satisfying (Ryan & Deci, 2000a) .  ... 
doi:10.19030/tlc.v13i2.9641 fatcat:mxtwlogiyfdihjprtas4agtaoi

Literature in language teaching: A recipe to maximize learning

Cagri Tugrul Mart
2018 L1-Educational Studies in Language and Literature  
The pedagogic rationale for welcoming literature in multiple learning settings lies in the claim that it is conducive to language learning.  ...  proficiency development.  ...  An integrated curriculum does not preclude the development of language proficiency; on the contrary, it maximizes learning experience (Barrette, Paesani, & Vinall, 2010) .  ... 
doi:10.17239/l1esll-2018.18.01.09 fatcat:maoegf4jfbaubicoykd354v3mq

Online Learning with Cumulative Oversampling: Application to Budgeted Influence Maximization [article]

Shatian Wang, Shuoguang Yang, Zhen Xu, Van-Anh Truong
2020 arXiv   pre-print
Combining CO with the oracle we design for the offline problem, our online learning algorithm simultaneously tackles budget allocation, parameter learning, and reward maximization.  ...  We propose a cumulative oversampling (CO) method for online learning.  ...  The agent needs to allocate the budget to T rounds as well as learning edge weights and maximizing cumulative reward.  ... 
arXiv:2004.11963v3 fatcat:jw5lcpspcbbolcqrgt6tbgbnfa

Vulnerable Medical Student Ecosystems: Transdisciplinary Learning Sciences Interventions, Maximizing Student Learning and Promoting Mental Health

Kevin Michael Watson
2021 European Journal of Social & Behavioural Sciences  
Finally, this article presents potential interventions, targeting the need for cultural change that may contribute to the creation of a more compassionate learning ecosystem to build the MS' mental resilience  ...  However, the literature shows otherwise; MS suffer debilitating anxiety and depression which worsen with the progression of their studies.  ...  system.  ... 
doi:10.15405/ejsbs.305 fatcat:gcvcbtysnnbzhlbt3fyrkvfh6q

Competence progress intrinsic motivation

Andrew Stout, Andrew G. Barto
2010 2010 IEEE 9th International Conference on Development and Learning  
Many researchers have suggested "intrinsically motivated" systems that receive internal reward for model learning progress, but for the most part this notion has not been applied with respect to skill  ...  Internal motivation systems take the place of (or augment) the external supervisory signals ubiquitous in traditional machine learning, providing to learning mechanisms an internally-generated reward signal  ...  ACKNOWLEDGMENTS Thanks to our colleagues in the Autonomous Learning Laboratory and to several anonymous reviewers for their feedback. A.  ... 
doi:10.1109/devlrn.2010.5578835 dblp:conf/icdl/StoutB10 fatcat:wioryiia7rcernvpesdiogymsi

Maximizing Interactivity in Online Learning: Moving Beyond Discussion Boards

Amber Dailey-Hebert
2018 Journal of Educators Online  
The e-portfolio provided a way for nurses to track, share, and showcase their skills, achievements, experience, competencies, and progress in continuous learning for both themselves, as evidence of learning  ...  The overarching recommendations to maximize learning for your students (and to save you time and effort) include: • Recommendation #1: Create opportunities for students to co-create and co-produce content  ...  Learning Management System: • What synchronous and asynchronous tools exist within your current LMS?  ... 
doi:10.9743/jeo.2018.15.3.8 fatcat:jztynanqsvgypm6dg2gjtlsh2m

Analysis of options for Maximizing Local Government internally generated Revenue in Nigeria

Moses Atakpa, Stephen Ocheni, Basil C. Nwankwo
2012 International Journal of Learning and Development  
This study examined the various options for maximizing internal revenue generation in the Nigerian local governments.  ...  The study suggested feasible and pragmatic ways to maximize internal revenue generation in the local governments.  ...  It is interesting to note that local government themselves are largely responsible for non-maximization of their internal revenue sources.  ... 
doi:10.5296/ijld.v2i5.2345 fatcat:cqtp4horsrhftb7oqdfnaaouam

Development of Student Worksheet-Based College E-Learning Through Edmodo to Maximize the Results of Learning and Motivation in Economic Mathematics Learning

Anton Nasrullah, Mira Marlina, Widya Dwiyanti
2018 International Journal of Emerging Technologies in Learning (iJET)  
The objective of the study was to develop a college student worksheet using Edmodo as a learning support tool and to know the effectiveness of peer tutor learning model applied in the learning of economic  ...  One class was used for the control group of 29 students; one for the exploratory group of 40 students.  ...  Acknowledgment The researcher would like to thank the Departement Mathematic Education of Bina Bangsa University and Departement Riset and Teknology for giving financial support. 8  ... 
doi:10.3991/ijet.v13i12.8636 fatcat:gyleea4cnvbeldrx2mpkdbmexy
« Previous Showing results 1 — 15 out of 69,685 results