Academic literature on the topic 'Online Sequential Learning From Preferences'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the lists of relevant articles, books, theses, conference reports, and other scholarly sources on the topic 'Online Sequential Learning From Preferences.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Journal articles on the topic "Online Sequential Learning From Preferences"
1
Ali, El mezouary, Hmedna Brahim, and Omar Baz. "An Unsupervised Method for Discovering How Does Learners' Progress Toward Understanding in MOOCs." International Journal of Innovative Technology and Exploring Engineering 10, no. 5 (March 30, 2021): 40–49. http://dx.doi.org/10.35940/ijitee.e8673.0310521.
Full textAbstract:
Massive Open Online Course (MOOC) seems to expand access to education and it present too many advantages as: democratization of learning, openness to all and accessibility on a large scale, etc. However, this new phenomenon of open learning suffers from the lack of personalization; it is not easy to identify learners’ characteristics because their heterogeneous masse. Following the increasing adoption of learning styles as personalization criteria, it is possible to make learning process easier for learners. In this paper, we extracted features from learners' traces when they interact with the MOOC platform in order to identify learning styles in an automatic way. For this purpose, we adopted the Felder-Silverman Learning Style Model (FSLSM) and used an unsupervised clustering method. Finally, this solution was implemented to clustered learners based on their level of preference for the sequential/global dimension of FSLSM. Results indicated that, first: k-means is the best performing algorithm when it comes to the identification of learning styles; second: the majority of learners show strong and moderate sequential learning style preferences.
2
Shkodina, Tatiana A. "Formation of an individual trajectory of online learning on the basis of cluster analysis." Journal Of Applied Informatics 18, no. 2 (March 31, 2023): 4–15. http://dx.doi.org/10.37791/2687-0649-2023-18-2-4-15.
Full textAbstract:
Justification for the relevance of developing an individual learning path in the field of online learning. The problems of forming an individual learning trajectory are analyzed. The main problem of personalization of learning from the point of view of the student is highlighted – the difficulty in finding the most appropriate sequence of studying educational objects that best suit their skills and preferences. It is concluded that the existing practices and methods of organizing a personalized educational process of courses in online learning are focused on the statistical characteristics of students that do not change during the study of an online course. Therefore, there is a need to develop a methodology for the formation of an individual learning path. The proposed approach allows us to consider the formation of recommendations as a dynamic process. An algorithm for the formation of an individual learning trajectory has been developed, which consists of a multi-criteria choice of a sequence of online courses at each moment of decision-making according to a given set of criteria and sequential mastering of skills. The choice of online courses is carried out using the cluster analysis method – k-means. Groups of clusters that meet the criteria of online courses have been identified. Each cluster consists of the closest objects – online courses. Based on these results, a sequential selection of online courses is made, using the available information about the user»s requirements and the skills that the learner needs to acquire. The purpose of developing for the formation of an individual learning trajectory is to provide students with the most appropriate sequence of learning objects in accordance with their skills and preferences.
3
Zheng, Yujia, Siyi Liu, Zekun Li, and Shu Wu. "Cold-start Sequential Recommendation via Meta Learner." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 5 (May 18, 2021): 4706–13. http://dx.doi.org/10.1609/aaai.v35i5.16601.
Full textAbstract:
This paper explores meta-learning in sequential recommendation to alleviate the item cold-start problem. Sequential recommendation aims to capture user's dynamic preferences based on historical behavior sequences and acts as a key component of most online recommendation scenarios. However, most previous methods have trouble recommending cold-start items, which are prevalent in those scenarios. As there is generally no side information in sequential recommendation task, previous cold-start methods could not be applied when only user-item interactions are available. Thus, we propose a Meta-learning-based Cold-Start Sequential Recommendation Framework, namely Mecos, to mitigate the item cold-start problem in sequential recommendation. This task is non-trivial as it targets at an important problem in a novel and challenging context. Mecos effectively extracts user preference from limited interactions and learns to match the target cold-start item with the potential user. Besides, our framework can be painlessly integrated with neural network-based models. Extensive experiments conducted on three real-world datasets verify the superiority of Mecos, with the average improvement up to 99%, 91%, and 70% in HR@10 over state-of-the-art baseline methods.
4
Bilmona, Hanafi. "Sequential Blended Teaching Materials: Scaffolding Non-English Language Learners’ Scientific Literacy Using Online Sources, Edpuzzle." PEJLaC: Pattimura Excellence Journal of Language and Culture 1, no. 1 (June 1, 2021): 26–33. http://dx.doi.org/10.30598/pejlac.v1.i1.pp26-33.
Full textAbstract:
This study aimed to explore information from 44 Students of Primary Teacher Training Education Department as the sample of research. The 40 items of statement in questionnaires designed from positive point of view and scaled in Likert from; strongly agree (5) to strongly disagree (1). By the content area of two basic research questions in regard to Preferences and benefit for students in applying a sequential blended learning material in teaching English. Data was analyzed in statistic descriptive to get meaning. The result found that, there was 4,22% of students respond to items of questionnaires to more agree and strongly agree to the statements of questionnaires. This total average number of the respond of 44 students were found in 4,6816 % in questionnaires that represented preferences (13 items of statements) and 3,7727% in average found in statement of questionnaires that represent benefits (27 items of statements), meant there was positive respond toward this teaching approach and fixing to the preferences of student and scaffolded students’ other related scientific literacy.
5
Jiang, Nan, Sheng Jin, Zhiyao Duan, and Changshui Zhang. "RL-Duet: Online Music Accompaniment Generation Using Deep Reinforcement Learning." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 01 (April 3, 2020): 710–18. http://dx.doi.org/10.1609/aaai.v34i01.5413.
Full textAbstract:
This paper presents a deep reinforcement learning algorithm for online accompaniment generation, with potential for real-time interactive human-machine duet improvisation. Different from offline music generation and harmonization, online music accompaniment requires the algorithm to respond to human input and generate the machine counterpart in a sequential order. We cast this as a reinforcement learning problem, where the generation agent learns a policy to generate a musical note (action) based on previously generated context (state). The key of this algorithm is the well-functioning reward model. Instead of defining it using music composition rules, we learn this model from monophonic and polyphonic training data. This model considers the compatibility of the machine-generated note with both the machine-generated context and the human-generated context. Experiments show that this algorithm is able to respond to the human part and generate a melodic, harmonic and diverse machine part. Subjective evaluations on preferences show that the proposed algorithm generates music pieces of higher quality than the baseline method.
6
Thaipisutikul, Tipajin. "An Adaptive Temporal-Concept Drift Model for Sequential Recommendation." ECTI Transactions on Computer and Information Technology (ECTI-CIT) 16, no. 2 (June 11, 2022): 222–36. http://dx.doi.org/10.37936/ecti-cit.2022162.248019.
Full textAbstract:
Recently, owing to the great advances in Web 2.0 and mobile devices, various online commercial services have emerged. Recommendation systems play an important role in dealing with abundant product information from massive numbers of online e-commerce transactions. Providing an accurate recommendation at the correct time to customers can contribute to a surge in business success. In this paper, an adaptive temporal-concept drift learning-based recommendation system, ATCRec, is developed for precisely tackling the sequential recommendation problem. We embed sequences of items into the latent spaces and learn both general preferences and sequential patterns concurrently via a recurrent neural network. Specifically, ATCRec captures dynamic changes in the temporal and concept drift contexts by modifying the gate units in a traditional recurrent neural network. The proposed model provides a unified and flexible network structure to learn and reveal the opaque variation of user preferences over time. We evaluate the robustness and performance of ATCRec on two real-world datasets, and the experimental results demonstrate that ATCRec consistently outperforms existing sequential recommendation approaches on various metrics. This indicates that integrating users' temporal information and concept drift variation through time are indispensable in improving the performance of recommendation systems.
7
Thaipisutikul, Tipajin. "An Adaptive Temporal-Concept Drift Model for Sequential Recommendation." ECTI Transactions on Computer and Information Technology (ECTI-CIT) 16, no. 2 (June 7, 2022): 221–35. http://dx.doi.org/10.37936/ecticit.2022162.248019.
Full textAbstract:
Recently, owing to the great advances in Web 2.0 and mobile devices, various online commercial services have emerged. Recommendation systems play an important role in dealing with abundant product information from massive numbers of online e-commerce transactions. Providing an accurate recommendation at the correct time to customers can contribute to a surge in business success. In this paper, an adaptive temporal-concept drift learning-based recommendation system, ATCRec, is developed for precisely tackling the sequential recommendation problem. We embed sequences of items into the latent spaces and learn both general preferences and sequential patterns concurrently via a recurrent neural network. Specifically, ATCRec captures dynamic changes in the temporal and concept drift contexts by modifying the gate units in a traditional recurrent neural network. The proposed model provides a unified and flexible network structure to learn and reveal the opaque variation of user preferences over time. We evaluate the robustness and performance of ATCRec on two real-world datasets, and the experimental results demonstrate that ATCRec consistently outperforms existing sequential recommendation approaches on various metrics. This indicates that integrating users' temporal information and concept drift variation through time are indispensable in improving the performance of recommendation systems.
8
Kan, Yirong, Kun Yue, Hao Wu, Xiaodong Fu, and Zhengbao Sun. "Online Learning of Parameters for Modeling User Preference Based on Bayesian Network." International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems 30, no. 02 (April 2022): 285–310. http://dx.doi.org/10.1142/s021848852250012x.
Full textAbstract:
By analyzing users’ behavior data for personalized services, most state-of-the-art methods for user preference modeling are often based on batch-mode machine learning algorithms, where all rating data are assumed to be available throughout the training process. However, data in the real world often arrives sequentially and user preference may change dynamically. The real-time characteristics of rating data make the algorithms for preference modeling challenging to suit real-world online applications. By the user preference model (UPM) based on Bayesian network with a latent variable (BNLV), uncertain relationships among relevant attributes of users, objects and ratings could be represented, in which user preference is represented by the latent variable. In this paper, we propose an online approach for parameter learning of UPM. Specifically, we first extend the classic Voting EM algorithm by using Bayesian estimation in terms of the situation with latent variables. Consequently, we propose the algorithm for learning parameters of UPM from few and sequentially-changing rating data to reflect the gradually changing preferences. Finally, we test the effectiveness of our proposed algorithm by conducting experiments on various datasets. Experimental results demonstrate the superiority of our method in various measurements.
9
Li, Zhao, Long Zhang, Chenyi Lei, Xia Chen, Jianliang Gao, and Jun Gao. "Attention with Long-Term Interval-Based Deep Sequential Learning for Recommendation." Complexity 2020 (July 13, 2020): 1–13. http://dx.doi.org/10.1155/2020/6136095.
Full textAbstract:
Modeling user behaviors as sequential learning provides key advantages in predicting future user actions, such as predicting the next product to purchase or the next song to listen to, for the purpose of personalized search and recommendation. Traditional methods for modeling sequential user behaviors usually depend on the premise of Markov processes, while recently recurrent neural networks (RNNs) have been adopted to leverage their power in modeling sequences. In this paper, we propose integrating attention mechanism into RNNs for better modeling sequential user behaviors. Specifically, we design a network featuring Attention with Long-term Interval-based Gated Recurrent Units (ALI-GRU) to model temporal sequences of user actions. Compared to previous works, our network can exploit the information of temporal dimension extracted by time interval-based GRU in addition to normal GRU to encoding user actions and has a specially designed matrix-form attention function to characterize both long-term preferences and short-term intents of users, while the attention-weighted features are finally decoded to predict the next user action. We have performed experiments on two well-known public datasets as well as a huge dataset built from real-world data of one of the largest online shopping websites. Experimental results show that the proposed ALI-GRU achieves significant improvement compared to state-of-the-art RNN-based methods. ALI-GRU is also adopted in a real-world application and results of the online A/B test further demonstrate its practical value.
10
Anisa, Anisa. "EFL Students’ Perceptions and Preferences of The Video Use as a Replacement for Traditional Lecture Method." IDEAS: Journal on English Language Teaching and Learning, Linguistics and Literature 10, no. 1 (June 10, 2022): 310–25. http://dx.doi.org/10.24256/ideas.v10i1.2656.
Full textAbstract:
This study aimed to investigate students’ perceptions and preferences of video use as a replacement for traditional lecture during e-learning in times of the COVID-19 pandemic. This study used two-staged mixed-method. The design of this study was sequential explanatory. This study used 30 closed-ended questionnaires and focus group discussions as the instrument to gain a rich understanding of students’ video experiences, perceptions and preferences. Data were collected from 108 EFL students through an online questionnaire using Google form and face-to-face focus group discussion. Results show that (1) Most of students showed positive perceptions of the use of video as a replacement for traditional lecture during e-learning in times of the COVID-19 Pandemic. It can be seen from the questionnaire results that 83.3% of students showed positive perception. (2) Students mostly preferred video-based learning if there is post-videos watching activities such as reviewing, re-explaining, question and answer, etc. Furthermore, the types of videos they preferred are instructor-created videos and animation videos from YouTube
Dissertations / Theses on the topic "Online Sequential Learning From Preferences"
1
Saha, Aadirupa. "Battle of Bandits: Online Learning from Subsetwise Preferences and Other Structured Feedback." Thesis, 2020. https://etd.iisc.ac.in/handle/2005/5184.
Full textAbstract:
The elicitation and aggregation of preferences is often the key to making better decisions. Be it a perfume company wanting to relaunch their 5 most popular fragrances, a movie recommender system trying to rank the most favoured movies, or a pharmaceutical company testing the relative efficacies of a set of drugs, learning from preference feedback is a widely applicable problem to solve. One can model the sequential version of this problem using the classical multiarmed-bandit (MAB) (e.g., Auer, 2002) by representing each decision choice as one bandit-arm, or more appropriately as a Dueling-Bandit (DB) problem (Yue \& Joachims, 2009). Although DB is similar to MAB in that it is an online decision making framework, DB is different in that it specifically models learning from pairwise preferences. In practice, it is often much easier to elicit information, especially when humans are in the loop, through relative preferences: `Item A is better than item B' is easier to elicit than its absolute counterpart: `Item A is worth 7 and B is worth 4'.
However, instead of pairwise preferences, a more general $k$-subset-wise preference model $(k \ge 2)$ is more relevant in various practical scenarios, e.g. recommender systems, search engines, crowd-sourcing, e-learning platforms, design of surveys, ranking in multiplayer games. Subset-wise preference elicitation is not only more budget friendly, but also flexible in conveying several types of feedback. For example, with subset-wise preferences, the learner could elicit the best item, a partial preference of the top 5 items, or even an entire rank ordering of a subset of items, whereas all these boil down to the same feedback over pairs (subsets of size 2). The problem of how to learn adaptively with subset-wise preferences, however, remains largely unexplored; this is primarily due to the computational burden of maintaining a combinatorially large, $O(n^k)$, size of preference information in general (for a decision problem with $n$ items and subsetsize $k$).
We take a step in the above direction by proposing ``Battling Bandits (BB)''---a new online learning framework to learn a set of optimal ('good') items by sequentially, and adaptively, querying subsets of items of size up to $k$ ($k\ge 2$). The preference feedback from a subset is assumed to arise from an underlying parametric discrete choice model, such as the well-known Plackett-Luce model, or more generally any random utility (RUM) based model. It is this structure that we leverage to design efficient algorithms for various problems of interest, e.g. identifying the best item, set of top-k items, full ranking etc., for both in PAC and regret minimization setting. We propose computationally efficient and (near-) optimal algorithms for above objectives along with matching lower bound guarantees. Interestingly this leads us to finding answers to some basic questions about the value of subset-wise preferences: Does playing a general $k$-set really help in faster information aggregation, i.e. is there a tradeoff between subsetsize-$k$ vs the learning rate? Under what type of feedback models? How do the performance limits (performance lower bounds) vary over different combinations of feedback and choice models? And above all, what more can we achieve through BB where DB fails?
We proceed to analyse the BB problem in the contextual scenario – this is relevant in settings where items have known attributes, and allows for potentially infinite decision spaces. This is more general and of practical interest than the finite-arm case, but, naturally, on the other hand more challenging. Moreover, none of the existing online learning algorithms extend straightforwardly to the continuous case, even for the most simple Dueling Bandit setup (i.e. when $k=2$). Towards this, we formulate the problem of ``Contextual Battling Bandits (C-BB)'' under utility based subsetwise-preference feedback, and design provably optimal algorithms for the regret minimization problem. Our regret bounds are also accompanied by matching lower bound guarantees showing optimality of our proposed methods. All our theoretical guarantees are corroborated with empirical evaluations.
Lastly, it goes without saying, that there are still many open threads to explore based on BB. These include studying different choice-feedback model combinations, performance objectives, or even extending BB to other useful frameworks like assortment selection, revenue maximization, budget-constrained bandits etc. Towards the end we will also discuss some interesting combinations of the BB framework with other, well-known, problems, e.g. Sleeping / Rotting Bandits, Preference based Reinforcement Learning, Learning on Graphs, Preferential Bandit-Convex-Optimization etc.
Book chapters on the topic "Online Sequential Learning From Preferences"
1
Guerin, Joshua T., Thomas E. Allen, and Judy Goldsmith. "Learning CP-net Preferences Online from User Queries." In Algorithmic Decision Theory, 208–20. Berlin, Heidelberg: Springer Berlin Heidelberg, 2013. http://dx.doi.org/10.1007/978-3-642-41575-3_16.
Full text
2
Stracke, Christian M., Aras Bozkurt, and Daniel Burgos. "Typologies of (Open) Online Courses and Their Dimensions, Characteristics and Relationships with Distributed Learning Ecosystems, Open Educational Resources, and Massive Open Online Courses." In Distributed Learning Ecosystems, 71–95. Wiesbaden: Springer Fachmedien Wiesbaden, 2023. http://dx.doi.org/10.1007/978-3-658-38703-7_5.
Full textAbstract:
AbstractThis chapter analyses the different typologies of online courses. First, we start with a reflection about the key terms of online learning, online courses, and distributed learning ecosystems (DLE). In our literature review, we cannot identify any existing typology framework for online courses. Consequently, we analyse and compare dimensions and categories of online courses from different sources: first, from the collected publications and studies identified in our literature review, second, from the current practices and platforms for online courses, and third, from standards for online courses, including the first international quality norm for online learning ISO/IEC 40180. As our key result, a framework proposal for the different typologies of online courses is developed based on these discussions and a comparison of several dimensions. The integration of our comparison results leads to the Typologies of Online Courses (TOC) framework with eight dimensions. The aim of the TOC framework is two-fold. First, it should support designers in the design, quality development, and evaluation of online courses. Second, it should enable learners to differentiate online courses according to the dimensions of these courses in comparison with their own preferences and demands. In the conclusion, an outlook on future research needs is provided. Finally, we come full circle and briefly discuss how (open) online courses and especially the two currently most important types, namely, Open Educational Resources (OER) and Massive Open Online Courses (MOOCs), can contribute to DLE and to addressing the general need for (equity and collaborative) education for all.
3
Jost, Patrick, and Monica Divitini. "From Paper to Online: Digitizing Card Based Co-creation of Games for Privacy Education." In Technology-Enhanced Learning for a Free, Safe, and Sustainable World, 178–92. Cham: Springer International Publishing, 2021. http://dx.doi.org/10.1007/978-3-030-86436-1_14.
Full textAbstract:
AbstractEducation is rapidly evolving from co-located settings to remote and online learning. However, many proven educational tools are designed for collaborative, co-located classroom work. Effective sketching and ideating tools, such as card-based workshop tools, cannot be applied in remote teaching.This paper explores how the paper-based card and playboard metaphor can be digitized for remote student co-creation via video call sessions. Therefore, a card-based toolkit for co-creating educational games is transformed into a digital representation for remote application. In a between-subject trial with two university student groups (n = 61), it is investigated how users perceive ideation/balancing support and applicability of the technology-enhanced card toolset compared to the paper-based variant. Both groups thereby created an analytic game concept for privacy education.The results remarkably revealed that remote co-creation using the technology-enhanced card and playboard in video call sessions was perceived as significantly more supportive for ideation and game concept balancing. Students also felt more confident to apply the digitized card toolset independently while being more satisfied with their created game concepts. The designed educational game concepts showed comparable patterns between the groups and disclosed the students’ preferences on how games for privacy education should be designed and when and where they would like to play them. Conclusively, design implications for digital card ideation toolsets were synthesized from the findings.
4
Ramponi, Giorgia. "Learning in the Presence of Multiple Agents." In Special Topics in Information Technology, 93–103. Cham: Springer International Publishing, 2022. http://dx.doi.org/10.1007/978-3-031-15374-7_8.
Full textAbstract:
AbstractReinforcement Learning (RL) has emerged as a powerful tool to solve sequential decision-making problems, where a learning agent interacts with an unknown environment in order to maximize its rewards. Although most RL real-world applications involve multiple agents, the Multi-Agent Reinforcement Learning (MARL) framework is still poorly understood from a theoretical point of view. In this manuscript, we take a step toward solving this problem, providing theoretically sound algorithms for three RL sub-problems with multiple agents: Inverse Reinforcement Learning (IRL), online learning in MARL, and policy optimization in MARL. We start by considering the IRL problem, providing novel algorithms in two different settings: the first considers how to recover and cluster the intentions of a set of agents given demonstrations of near-optimal behavior; the second aims at inferring the reward function optimized by an agent while observing its actual learning process. Then, we consider online learning in MARL. We showed how the presence of other agents can increase the hardness of the problem while proposing statistically efficient algorithms in two settings: Non-cooperative Configurable Markov Decision Processes and Turn-based Markov Games. As the third sub-problem, we study MARL from an optimization viewpoint, showing the difficulties that arise from multiple function optimization problems and providing a novel algorithm for this scenario.
5
Williams, Geoffrey Alan. "Understanding the Preferences for Online Learning." In Advancing Innovation and Sustainable Outcomes in International Graduate Education, 194–208. IGI Global, 2021. http://dx.doi.org/10.4018/978-1-7998-5514-9.ch012.
Full textAbstract:
Online learning is promoted by the Malaysian Government as a key element in the Higher Education Blueprint 2015-25 (Shift 9: Globalized Online Learning), but research in the Malaysian context is very underdeveloped. This chapter aims to fill part of this gap with a simple analysis of online Master of Business Administration (MBA) courses to examine the appetite and preferences of actual and potential MBA students for online learning. Using data from local and international students studying on MBA programs in Malaysia, the authors show that the MBA students in their sample still have a largely instrumental view of the value drivers of their study programs. The key factors identified by the largest number of groups were facilities, price, certificate authenticity, duration, and flexibility of course times.
6
Hall, Andrea. "Designing Culturally Appropriate E-Learning for Learners from an Arabic Background." In Cases on Globalized and Culturally Appropriate E-Learning, 94–113. IGI Global, 2011. http://dx.doi.org/10.4018/978-1-61520-989-7.ch005.
Full textAbstract:
Case studies on adult online learners in professional development courses in an Omani context found that cultural preferences had a significant impact on learning success. It was found that their preferences in the development of learning communities, for face-to-face needs, in online course flexibility, and interdependent learning were not accounted for in the learning design. Therefore, the problem identified was: how can learning be designed that accounts for culture in the design of learning for those from an Arabic cultural background, as in Oman? The research provided a solution in the form of design guidelines. These can be used as a practical and useful means for teachers and educators in designing online courses that are culturally compatible with the learning preferences in this context in the Sultanate of Oman.
7
Sharma, Meenakshi, and Alka Dwivedi. "Relationship Between Online Learning Environments and Student Behaviour." In Technology Training for Educators From Past to Present, 239–49. IGI Global, 2022. http://dx.doi.org/10.4018/978-1-6684-4083-4.ch012.
Full textAbstract:
The coronavirus disease originated from the city of Wuhan in late 2019. Colleges and schools faced various challenges, but they accepted the fact and soon went online and learning continued. This study makes an attempt to understand student behaviour towards the online learning environment and to measure the relationship between the various online learning environment factors and student behaviour. Primary data of 500 students were collected using stratified random sampling from different schools and universities with the help of a questionnaire. Data were analysed using Karl Pearson correlation. Results show a weak correlation between online learning factors and student behaviour towards learning. The exclusivity of this study is that it has attempted to fill the wide gaps existing in the field of online learning. This research work confirms the importance of student preferences in analysing and measuring the experiences of students and identifies their significant effect in online learning especially in higher education.
8
Kumar Mitra, Nilesh. "New Updates in Online Learning." In New Updates in E-Learning [Working Title]. IntechOpen, 2022. http://dx.doi.org/10.5772/intechopen.102576.
Full textAbstract:
During the COVID-19 pandemic, a rapid transformation happened overnight in the teaching-learning strategy in primary, secondary and tertiary education. All educators started using web-conferencing tools as principal element of online learning. However, in spite of health concerns among the pandemic situation, strong student preferences towards returning back to face-to-face or hybrid mode brought challenges to the effectiveness of online learning. Students cite many reasons for dropping out of online courses. Increased workload and poor organization of remote learning have been found to be the principal reason for the students’ dissatisfaction. The orientation of online learning needs alignment towards the principle of course design along with the flexibility to attain the instructional goals, objectives, and outcomes. Sophisticated technology often makes online and even hybrid course design to change track from well-designed pedagogy leading to loss of functional relevance for the students. Instructors should be flexible and employ multiple strategies to improve online learning experiences in both asynchronous and synchronous learning environment. Studies have proved that using the best practice of the alignment of learning outcome, online learning activities and repeated online knowledge-checks foster student motivation towards the completion of online courses.
9
Zhao, Jinjing. "L2 Languaging in a Massively Multiplayer Online Game." In Computer-Assisted Language Learning, 855–72. IGI Global, 2019. http://dx.doi.org/10.4018/978-1-5225-7663-1.ch040.
Full textAbstract:
This paper examines L2 learner variations in the context of massively multiplayer online games (MMOGs). MMOGs have gained much attention among CALL researchers because this particular game genre is perceived to promote informal, contextualized interaction in a learner's target language, including interaction with native speakers. However, there is little research on differences between L2 learners in terms of how they engage in language learning and use in the context of gameplay. Drawing on data from questionnaires, interviews, gaming sessions, and gaming journals, this paper argues that affordances of MMOGs must be understood in relation to the learner's history, ability, and preference within the social context of game play; L2 learners engage with various game discourses that align with their preferences of game play and goals of language learning. In closing, the paper discusses procedural challenges in conducting research on MMOGs and similar gaming contexts.
10
Kale, Archana P., Shefali P. Sonavane, Shashwati P. Kale, and Aditi R. Wade. "Multimodal Genetic Optimized Feature Selection for Online Sequential Extreme Learning Machine." In Artificial Intelligence and Natural Algorithms, 250–60. BENTHAM SCIENCE PUBLISHERS, 2022. http://dx.doi.org/10.2174/9789815036091122010017.
Full textAbstract:
Extreme learning machine (ELM) is a rapid classifier evolved for batch learning mode unsuitable for sequential input. Retrieving data from the new inventory leads to a time-extended process. Therefore, online sequential extreme learning machine (OSELM) algorithms were proposed by Liang et al.. The OSELM is able to handle the sequential input by reading data 1 by 1 or chunk by chunk mode. The overall system generalization performance may devalue because of the amalgamation of the random initialization of OS-ELM and the presence of redundant and irrelevant features. To resolve the said problem, this paper proposes a correspondence multimodal genetic optimized feature selection paradigm for sequential input (MG-OSELM) for radial basis function by using clinical datasets. For performance comparison, the proposed paradigm is implemented and evaluated for ELM, multimodal genetic optimized for ELM classifier (MG-ELM), OS-ELM, MG-OSELM. Experimental results are calculated and analysed accordingly. The comparative results analysis illustrates that MG-ELM provides 10.94% improved accuracy with 43.25% features compared to ELM.
Conference papers on the topic "Online Sequential Learning From Preferences"
1
Wu, Bo, Wen-Huang Cheng, Yongdong Zhang, Qiushi Huang, Jintao Li, and Tao Mei. "Sequential Prediction of Social Media Popularity with Deep Temporal Context Networks." In Twenty-Sixth International Joint Conference on Artificial Intelligence. California: International Joint Conferences on Artificial Intelligence Organization, 2017. http://dx.doi.org/10.24963/ijcai.2017/427.
Full textAbstract:
Prediction of popularity has profound impact for social media, since it offers opportunities to reveal individual preference and public attention from evolutionary social systems. Previous research, although achieves promising results, neglects one distinctive characteristic of social data, i.e., sequentiality. For example, the popularity of online content is generated over time with sequential post streams of social media. To investigate the sequential prediction of popularity, we propose a novel prediction framework called Deep Temporal Context Networks (DTCN) by incorporating both temporal context and temporal attention into account. Our DTCN contains three main components, from embedding, learning to predicting. With a joint embedding network, we obtain a unified deep representation of multi-modal user-post data in a common embedding space. Then, based on the embedded data sequence over time, temporal context learning attempts to recurrently learn two adaptive temporal contexts for sequential popularity. Finally, a novel temporal attention is designed to predict new popularity (the popularity of a new user-post pair) with temporal coherence across multiple time-scales. Experiments on our released image dataset with about 600K Flickr photos demonstrate that DTCN outperforms state-of-the-art deep prediction algorithms, with an average of 21.51% relative performance improvement in the popularity prediction (Spearman Ranking Correlation).
2
Myers, Vivek, Erdem Bıyık, and Dorsa Sadigh. "Active Reward Learning from Online Preferences." In 2023 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2023. http://dx.doi.org/10.1109/icra48891.2023.10160439.
Full text
3
Vlaski, Stefan, Hermina P. Maretic, Roula Nassif, Pascal Frossard, and Ali H. Sayed. "ONLINE GRAPH LEARNING FROM SEQUENTIAL DATA." In 2018 IEEE Data Science Workshop (DSW). IEEE, 2018. http://dx.doi.org/10.1109/dsw.2018.8439913.
Full text
4
Luo, Yong, Tongliang Liu, Yonggang Wen, and Dacheng Tao. "Online Heterogeneous Transfer Metric Learning." In Twenty-Seventh International Joint Conference on Artificial Intelligence {IJCAI-18}. California: International Joint Conferences on Artificial Intelligence Organization, 2018. http://dx.doi.org/10.24963/ijcai.2018/350.
Full textAbstract:
Distance metric learning (DML) has been demonstrated to be successful and essential in diverse applications. Transfer metric learning (TML) can help DML in the target domain with limited label information by utilizing information from some related source domains. The heterogeneous TML (HTML), where the feature representations vary from the source to the target domain, is general and challenging. However, current HTML approaches are usually conducted in a batch manner and cannot handle sequential data. This motivates the proposed online HTML (OHTML) method. In particular, the distance metric in the source domain is pre-trained using some existing DML algorithms. To enable knowledge transfer, we assume there are large amounts of unlabeled corresponding data that have representations in both the source and target domains. By enforcing the distances (between these unlabeled samples) in the target domain to agree with those in the source domain under the manifold regularization theme, we learn an improved target metric. We formulate the problem in the online setting so that the optimization is efficient and the model can be adapted to new coming data. Experiments in diverse applications demonstrate both effectiveness and efficiency of the proposed method.
5
Pereira, Fabiola S. F., Gina M. B. Oliveira, and João Gama. "User Preference Dynamics on Evolving Social Networks - Learning, Modeling and Prediction." In XXV Simpósio Brasileiro de Sistemas Multimídia e Web. Sociedade Brasileira de Computação - SBC, 2019. http://dx.doi.org/10.5753/webmedia_estendido.2019.8129.
Full textAbstract:
The preferences adopted by individuals are constantly modified as these are driven by new experiences, natural life evolution and, mainly, influence from friends. Studying these temporal dynamics of user preferences has become increasingly important for personalization tasks. Online social networks contain rich information about social interactions and relations, becoming essential source of knowledge for the understanding of user preferences evolution. In this thesis, we investigate the interplay between user preferences and social networks over time. We use temporal networks to analyze the evolution of social relationships and propose strategies to detect changes in the network structure based on node centrality. Our findings show that we can predict user preference changes by just observing how her social network structure evolves over time.
6
Sioson, Irish Chan. "Attitudes of Thai English Learners towards Online Learning of Speaking." In 16th Education and Development Conference. Tomorrow People Organization, 2021. http://dx.doi.org/10.52987/edc.2021.003.
Full textAbstract:
ABSTRACT A recent challenge in the field of education has been met as a shift to online classes from traditional face-to-face classes has been attributed to the COVID-19 pandemic. Hence, certain issues arise from such a sudden shift to an online learning environment, especially for those who have been mainly (or for others, solely) taught in a face-to-face setting. This paper aimed to determine the attitudes of Thai English learners towards online learning of speaking. The study involved fifty-four fourth year English majors in a university in southern Thailand. A survey questionnaire was developed to collect data. It consisted of a 5- point Likert scale asking for the students' level of agreement with statements and open-ended questions. The results show that the teacher being perceived as supportive and the students having a positive feeling when they had a stable Internet connection were the two areas that had the highest mean scores. On the other hand, being given enough opportunities to interact with classmates and preferring to participate in discussions using video (with microphone and video on) had the lowest mean scores. Moreover, the learners’ comments provided insights into their attitudes toward online learning in terms of preferences and challenges. Implications for teaching are then drawn from the results. KEYWORDS: attitudes, online learning, speaking
7
Zhou, Xiao, Danyang Liu, Jianxun Lian, and Xing Xie. "Collaborative Metric Learning with Memory Network for Multi-Relational Recommender Systems." In Twenty-Eighth International Joint Conference on Artificial Intelligence {IJCAI-19}. California: International Joint Conferences on Artificial Intelligence Organization, 2019. http://dx.doi.org/10.24963/ijcai.2019/619.
Full textAbstract:
The success of recommender systems in modern online platforms is inseparable from the accurate capture of users' personal tastes. In everyday life, large amounts of user feedback data are created along with user-item online interactions in a variety of ways, such as browsing, purchasing, and sharing. These multiple types of user feedback provide us with tremendous opportunities to detect individuals' fine-grained preferences. Different from most existing recommender systems that rely on a single type of feedback, we advocate incorporating multiple types of user-item interactions for better recommendations. Based on the observation that the underlying spectrum of user preferences is reflected in various types of interactions with items and can be uncovered by latent relational learning in metric space, we propose a unified neural learning framework, named Multi-Relational Memory Network (MRMN). It can not only model fine-grained user-item relations but also enable us to discriminate between feedback types in terms of the strength and diversity of user preferences. Extensive experiments show that the proposed MRMN model outperforms competitive state-of-the-art algorithms in a wide range of scenarios, including e-commerce, local services, and job recommendations.
8
Jin, Jian, Ying Liu, Ping Ji, and Richard Fung. "Design Preference Centered Review Recommendation: A Similarity Learning Approach." In ASME 2011 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. ASMEDC, 2011. http://dx.doi.org/10.1115/detc2011-48181.
Full textAbstract:
The rise of e-commerce websites like Amazon and Alibaba is changing the way how designers seek information to identify customer preferences in product design. From the feedbacks posted by consumers, either positive or negative, product designers can monitor the trend of consumers’ perception with respect to their product offerings and make efforts to improve accordingly. Starting from feature extraction from review documents, existing methods in identifying helpful online reviews regard the helpfulness prediction problem as a regression or classification problem and have not considered the relationship between customer reviews. Also, these approaches only consider the online helpfulness voting ratio or a unified helpfulness rating as the gold criteria for helpfulness evaluation and neglect various personal preferences from product designers. Therefore, in this paper, the focus is on how to predict reviews’ helpfulness by taking into account the personal preferences from both reviewers and designers. We start to analyze review helpfulness from both a generic aspect and a personal preference aspect. Classification methods and the proposed review similarity learning approach are utilized to estimate from the generic angle of helpfulness, while nearest neighbourhood based methods are adopted to reflect concerns from personal perspectives. Finally, a regression algorithm is called upon to predict review helpfulness based on the inputs from both aspects. Our experimental study, using a large quantity of review data crawled from Amazon and real ratings from product designers demonstrates the effectiveness of our approach and it opens a possibility for customized helpful review routing.
9
Li, Chang, and Maarten de Rijke. "Cascading Non-Stationary Bandits: Online Learning to Rank in the Non-Stationary Cascade Model." In Twenty-Eighth International Joint Conference on Artificial Intelligence {IJCAI-19}. California: International Joint Conferences on Artificial Intelligence Organization, 2019. http://dx.doi.org/10.24963/ijcai.2019/396.
Full textAbstract:
Non-stationarity appears in many online applications such as web search and advertising. In this paper, we study the online learning to rank problem in a non-stationary environment where user preferences change abruptly at an unknown moment in time. We consider the problem of identifying the K most attractive items and propose cascading non-stationary bandits, an online learning variant of the cascading model, where a user browses a ranked list from top to bottom and clicks on the first attractive item. We propose two algorithms for solving this non-stationary problem: CascadeDUCB and CascadeSWUCB. We analyze their performance and derive gap-dependent upper bounds on the n-step regret of these algorithms. We also establish a lower bound on the regret for cascading non-stationary bandits and show that both algorithms match the lower bound up to a logarithmic factor. Finally, we evaluate their performance on a real-world web search click dataset.
10
Feldhammer-Kahr, Martina, Stefan Dreisiebner, Martin Arendasy, and Manuela Paechter. "ONE MONTH BEFORE THE PANDEMIC: STUDENTS’ PREFERENCES FOR FLEXIBLE LEARNING AND WHAT WE CAN LEARN." In International Psychological Applications Conference and Trends. inScience Press, 2021. http://dx.doi.org/10.36315/2021inpact039.
Full textAbstract:
"Flexible learning has been associated with e-learning, even before the COVID-19 pandemic. Flexible learning gives the students large degrees of freedom to learn what, when, how and where they want. The aim of this study was to evaluate students’ preferences in e-learning and traditional classroom teaching, and was conducted from October 2019 to January 2020. Students from four courses were assigned randomly to two groups, an online and a classroom group. The study included two phases: three lectures by the lecturer (podcasts vs. classroom) and seven classroom units with student presentations and discussions. Performance and different personal characteristics and attitudes of 93 students were examined. Knowledge on the course topic was measured before the first lecture took place (t1), after the three lectures (t2) and after the following seven units (t3). Statistical analyses found no performance differences between the two groups (online/classroom); this held true for all three points in time. All students appreciated the opportunity of an intermediate exam at t2 (a change in comparison to former courses on the topic). Qualitative data showed that students felt a need for interaction with their colleagues and the lecturer, which they decided could be better fulfilled in the classroom, whereas the flexible learning setting had advantages for the exam preparation (e.g. repeating listening to the podcasts, taking breaks and learning tempo). Students’ arguments fit well to previous literature. Altogether, the study gives valuable insights into the didactic design of flexible learning."