550 Hits in 9.0 sec

The Deep Bootstrap Framework: Good Online Learners are Good Offline Generalizers [article]

Preetum Nakkiran, Behnam Neyshabur, Hanie Sedghi
2021 arXiv   pre-print
If the gap (2) is universally small, this reduces the problem of generalization in offline learning to the problem of optimization in online learning.  ...  We propose a new framework for reasoning about generalization in deep learning.  ...  PN also supported in part by a Google PhD Fellowship, the Simons Investigator Awards of Boaz Barak and Madhu Sudan, and NSF Awards under grants CCF 1565264, CCF 1715187 and IIS 1409097.  ... 
arXiv:2010.08127v2 fatcat:d5vmn2sbxzamzhmb6bue6xhney

Acme: A Research Framework for Distributed Reinforcement Learning [article]

Matt Hoffman, Bobak Shahriari, John Aslanides, Gabriel Barth-Maron, Feryal Behbahani, Tamara Norman, Abbas Abdolmaleki, Albin Cassirer, Fan Yang, Kate Baumli, Sarah Henderson, Alex Novikov (+8 others)
2020 arXiv   pre-print
To this end we are releasing baseline implementations of various algorithms, created using our framework.  ...  Deep reinforcement learning has led to many recent-and groundbreaking-advancements.  ...  ., 2017) and TF-Agents (Sergio Guadarrama, 2018) are both examples of established deep RL frameworks written in TensorFlow 1.X.  ... 
arXiv:2006.00979v1 fatcat:66owwgu3yzby3mfrxqtno6fg4i

LeSSA: A Unified Framework based on Lexicons and Semi-Supervised Learning Approaches for Textual Sentiment Classification

Jawad Khan, Young-Koo Lee
2019 Applied Sciences  
SA aims to classify the online unstructured user-generated contents (UUGC) into positive and negative classes.  ...  (c) classification fusion, whereby the predictions from numerous learners are confluences to determine final sentiment polarity based on majority voting, and (d) practicality, that is, we validate our  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/app9245562 fatcat:adzlvshbmbfklew457auwrh7ue

Burn After Reading: Online Adaptation for Cross-domain Streaming Data [article]

Luyu Yang, Mingfei Gao, Zeyuan Chen, Ran Xu, Abhinav Shrivastava, Chetan Ramaiah
2021 arXiv   pre-print
Thus we propose an online framework that "burns after reading", i.e. each online sample is immediately deleted after it is processed.  ...  Further, to fully exploit the valuable discrepancies among the diverse combinations, we employ the training strategy of multiple learners with co-supervision.  ...  Acknowledgement We thank Caiming Xiong for his valuable insights into the project, Junnan Li and Shu Zhang for the help to improve the experiments.  ... 
arXiv:2112.04345v1 fatcat:c2nyiq7vxnai7cqydo5mqzsd7u

Wearable Learning for Healthy Ageing through Creative Learning: A Conceptual Framework in the project "Fitness MOOC" (fMOOC)

Ilona Buchem, Jorn Kreutel, Agathe Merceron, Marten Haesner, Anika Steinert
2015 Interaction Design and Architecture(s)  
In this paper we present the conceptual framework and the architecture of wearable-technology enhanced learning for healthy ageing as part of an R&D project called "Fitness MOOC - interaction of seniors  ...  The project aims at developing a wearable-technology enhanced learning solution combining the MOOC (Massive Open Online Course) approach with embodied and creative learning experience with support of activity  ...  The online components are embedded in the fMOOC platform and include digital content and online communication with other learners.  ... 
doaj:422deef9deab4b179cbf1a08306a09c1 fatcat:6ykhopoxjjhr7hyzmszfbksiwu

How Flow Experience and Self-Efficacy Define Students' Online Learning Intentions: View From Task Technology Fit (Framework)

Hai Huang, Yong Wang
2022 Frontiers in Psychology  
However, the learners' self-efficacy is significant enough to map learners' intentions to use an online environment for learning.  ...  Findings conclude that flow experience is the most critical factor to define learners' perceived TTF in the case of an online learning experience.  ...  In addition, for BSG providers, technology support needs a self-assistive environment and tools, which can help learners to practice BSG in the extended mixed approach, where online and offline BSG usage  ... 
doi:10.3389/fpsyg.2022.835328 pmid:35369249 pmcid:PMC8965651 fatcat:fza6ddr2qngcve526qt7kftlbe

AutonoML: Towards an Integrated Framework for Autonomous Machine Learning [article]

David Jacob Kedziora and Katarzyna Musial and Bogdan Gabrys
2022 arXiv   pre-print
However, these ambitions are unlikely to be achieved in a robust manner without the broader synthesis of various mechanisms and theoretical frameworks, which, at the present time, remain scattered across  ...  Ultimately, we conclude that the notion of architectural integration deserves more discussion, without which the field of automated ML risks stifling both its technical advantages and general uptake.  ...  Two examples of this process are k-fold cross-validation and out-of-bootstrap validation.  ... 
arXiv:2012.12600v2 fatcat:6rj4ubhcjncvddztjs7tql3itq

Curriculum Learning for Reinforcement Learning Domains: A Framework and Survey [article]

Sanmit Narvekar and Bei Peng and Matteo Leonetti and Jivko Sinapov and Matthew E. Taylor and Peter Stone
2020 arXiv   pre-print
Finally, we use our framework to find open problems and suggest directions for future RL curriculum learning research.  ...  Despite many advances over the past three decades, learning in many domains still requires a large amount of interaction with the environment, which can be prohibitively expensive in realistic scenarios  ...  Part of this work has taken place in the Learning Agents Research Group (LARG) at the Artificial Intelligence Laboratory, The University of Texas at Austin. LARG re-  ... 
arXiv:2003.04960v2 fatcat:iacmqeb7jjeezpo27jsnzuqb7u

An Introduction to Advanced Machine Learning : Meta Learning Algorithms, Applications and Promises [article]

Farid Ghareh Mohammadi, M. Hadi Amini, Hamid R. Arabnia
2019 arXiv   pre-print
These algorithms, however, are not tailored for solving emerging learning problems. One of the important issues caused by online data is lack of sufficient samples per class.  ...  Majority of optimization algorithms that have been introduced in [1, 2] guarantee the optimal performance of supervised learning, given offline and discrete data, to deal with curse of dimensionality (  ...  Meta learner also is one of the bootstrap algorithms which learn data by sampling given data set and generating different data sets.  ... 
arXiv:1908.09788v1 fatcat:qujten7zzzbd7laazhymnfw2yi

A machine learning framework to improve effluent quality control in wastewater treatment plants

Dong Wang, Sven Thunéll, Ulrika Lindberg, Lili Jiang, Johan Trygg, Mats Tysklind, Nabil Souihi
2021 Science of the Total Environment  
The framework could also be applied to other parameters in WWTPs and industrial processes in general if sufficient high-resolution data are available.  ...  The framework consists of Random Forest (RF) models, Deep Neural Network (DNN) models, Variable Importance Measure (VIM) analyses, and Partial Dependence Plot (PDP) analyses, and uses a novel approach  ...  Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2021.147138.  ... 
doi:10.1016/j.scitotenv.2021.147138 pmid:34088065 fatcat:frlvnldeqfb5jajczudbdfovoi

Item Response Theory – A Statistical Framework for Educational and Psychological Measurement [article]

Yunxiao Chen, Xiaoou Li, Jingchen Liu, Zhiliang Ying
2021 arXiv   pre-print
Possible future directions of IRT are discussed from the perspective of statistical learning.  ...  The IRT models are latent factor models tailored to the analysis, interpretation, and prediction of individuals' behaviors in answering a set of measurement items that typically involve categorical response  ...  Offline estimation. We first consider the offline setting for the estimation of person parameters, assuming that the item parameters are known.  ... 
arXiv:2108.08604v1 fatcat:4qkgd6wc4zfc3fn6zxy5mq2xsm

A Supervised Machine Learning Classification Framework for Clothing Products' Sustainability

Chloe Satinet, François Fouss
2022 Sustainability  
The resulting model provides rapid environmental feedback on a variety of clothing products with the limited data available to online retailers.  ...  Indeed, there are a variety of confusing sustainability certifications and few labels capturing the overall environmental impact of products, as the existing procedures for assessing the environmental  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/su14031334 fatcat:hyfjmmtazvak3aajvxwkeb7xhm

Meta-Gradient Reinforcement Learning with an Objective Discovered Online [article]

Zhongwen Xu, Hado van Hasselt, Matteo Hessel, Junhyuk Oh, Satinder Singh, David Silver
2020 arXiv   pre-print
Over time, this allows the agent to learn how to learn increasingly effectively. Furthermore, because the objective is discovered online, it can adapt to changes over time.  ...  We demonstrate that the algorithm discovers how to address several important issues in RL, such as bootstrapping, non-stationarity, and off-policy learning.  ...  Precup for their comments and suggestions on the paper.  ... 
arXiv:2007.08433v1 fatcat:ljl2ig64rffmphbluh24zpceoq

Immersive Experience during Covid-19: The Mediator Role of Alternative Assessment in Online Learning Environment

Mohd Hanafi Azman Ong, Norazlina Mohd Yasin, Nur Syafikah Ibrahim
2021 International Journal of Interactive Mobile Technologies  
The study particularly examines how various factors are associated with students' online learning experience, particularly during the pandemic.  ...  An online survey of 312 respondents who used the Blackboard online learning platform was conducted, and a PLS-SEM analysis indicated that an alternative assessment mediated the relationship of learning  ...  Blended learning integrates online and offline instruction to foster deep and meaningful learning while maintaining the principles associated with conventional academic institutions [2] .  ... 
doi:10.3991/ijim.v15i18.24541 fatcat:wxxep3vkifbfjhgj4vxe3ujqju

Offline Data-Driven Evolutionary Optimization Using Selective Surrogate Ensembles

Handing Wang, Yaochu Jin, Chaoli Sun, John Doherty
2018 IEEE Transactions on Evolutionary Computation  
Such problems are known as offline data-driven optimization problems.  ...  This paper proposes a new offline data-driven evolutionary algorithm to make the full use of the offline data to guide the search.  ...  Before running the optimizer (a canonical EA), offline data is created, from which T subsets (S 1 , S 2 ,..., S T ) are generated using bootstrap. Then, T models (M 1 , M 2 ,...  ... 
doi:10.1109/tevc.2018.2834881 fatcat:auy7y4siq5einpetnklvglbeoe
« Previous Showing results 1 — 15 out of 550 results