197,763 Hits in 2.6 sec

Deep Knowledge Tracing with Transformers [chapter]

Shi Pu, Michael Yudelson, Lu Ou, Yuchi Huang
2020 Lecture Notes in Computer Science  
In this work, we propose a Transformer-based model to trace students' knowledge acquisition.  ...  Our approach outperforms the state-of-theart methods in the literature by roughly 10% in AUC with frequently used public datasets.  ...  One of such extensions is Deep Knowledge Tracing (DKT). The first DKT [6] adopted the Recurrent Neural Network (RNN) architecture from the deep learning community.  ... 
doi:10.1007/978-3-030-52240-7_46 fatcat:5yeu3zff3raw5gpklrlqvck7rm

Towards an Appropriate Query, Key, and Value Computation for Knowledge Tracing [article]

Youngduck Choi, Youngnam Lee, Junghyun Cho, Jineon Baek, Byungsoo Kim, Yeongmin Cha, Dongmin Shin, Chan Bae, Jaewe Heo
2020 arXiv   pre-print
In this paper, we propose a novel Transformer based model for knowledge tracing, SAINT: Separated Self-AttentIve Neural Knowledge Tracing.  ...  To the best of our knowledge, this is the first work to suggest an encoder-decoder model for knowledge tracing that applies deep self-attentive layers to exercises and responses separately.  ...  Transformer based Variants of Deep Knowledge Tracing Models In order to find a deep attention mechanism effective for knowledge tracing, we conduct extensive experiments on different ways to construct  ... 
arXiv:2002.07033v4 fatcat:4hezbch5ojdudd7n7miij5xmpa

Application of Deep Self-Attention in Knowledge Tracing [article]

Junhao Zeng, Qingchun Zhang, Ning Xie, Bochun Yang
2021 arXiv   pre-print
This paper proposed Deep Self-Attentive Knowledge Tracing (DSAKT) based on the data of PTA, an online assessment system used by students in many universities in China, to help these students learn more  ...  The intelligent tutoring system must model learners' mastery of the knowledge before providing feedback and advices to learners, so one class of algorithm called "knowledge tracing" is surely important  ...  In this paper we improved SAKT and proposed an encoderdecoder based model named Deep Self-Attentive Knowledge Tracing (DSAKT).  ... 
arXiv:2105.07909v2 fatcat:ubp22jwoeba3taso5rkwnur2su

Deep Open Snake Tracker for Vessel Tracing [article]

Li Chen, Wenjin Liu, Niranjan Balu, Mahmud Mossa-Basha, Thomas S. Hatsukami, Jenq-Neng Hwang, Chun Yuan
2021 arXiv   pre-print
Vessel tracing by modeling vascular structures in 3D medical images with centerlines and radii can provide useful information for vascular health.  ...  We propose here a deep learning based open curve active contour model (DOST) to trace vessels in 3D images. Initial curves were proposed from a centerline segmentation neural network.  ...  existing models with either human or machine knowledge.  ... 
arXiv:2107.09049v1 fatcat:vgq4yc3llbfsffgrrq6sq6vaka

Knowledge engineering requirements in derivational analogy [chapter]

Pádraig Cunningham, Donal Finn, Seán Slattery
1994 Lecture Notes in Computer Science  
In the adaptation process this reasoning trace is reinstantiated in the context of the new target case; this requires a strong domain model and the encoding of problem solving knowledge.  ...  In DA the case representation contains a trace of the reasoning process involved in producing the solution for that case.  ...  Transformation adaptation requires retrieval based on more abstract features and needs access to a deep domain model.  ... 
doi:10.1007/3-540-58330-0_90 fatcat:jn2hgsrmkzeozaw67vjssvho6q

Applying general markup knowledge to analyze ionograms of various ionosondes

Vladimir Mochalov, Anastasia Mochalova, A. Dmitriev, G. Vodinchar, N. Salikhov, Z. Rakhmonov
2020 E3S Web of Conferences  
On the basis of reference markings from two ionosondes, deep neural networks were trained to highlight reflection traces from different layers of the ionosphere.  ...  In order to improve the quality of recognition of ionograms, the use of general knowledge about the reference marking of ionograms at various points of installation of ionosondes of the same type is considered  ...  expert operators involved in the processing and interpretation of ionograms, the results of which served as the basis for creating training data for the ionogram recognition system based on the use of deep  ... 
doi:10.1051/e3sconf/202019603002 fatcat:ya65aazqmjhd7dvqnk36sozs5i

Deep Knowledge Tracing with Learning Curves [article]

Shanghui Yang, Mengxia Zhu, Xuesong Lu
2021 arXiv   pre-print
Knowledge tracing (KT) has recently been an active research area of computational pedagogy.  ...  Based on this theory, we propose a Convolution-Augmented Knowledge Tracing (CAKT) model in this paper.  ...  Knowledge Tracing with Deep Learning The Deep Knowledge Tracing (DKT) model [28] first applies deep learning on the KT task.  ... 
arXiv:2008.01169v2 fatcat:ogcrwrc26nahrjfguxawxlapki

Semantics of the Black-Box: Can knowledge graphs help make deep learning systems more interpretable and explainable? [article]

Manas Gaur, Keyur Faldu, Amit Sheth
2020 arXiv   pre-print
The recent series of innovations in deep learning (DL) have shown enormous potential to impact individuals and society, both positively and negatively.  ...  This article demonstrates how knowledge, provided as a knowledge graph, is incorporated into DL methods using knowledge-infused learning, which is one of the strategies.  ...  On the other hand, deep infusion of knowledge is a paradigm that couples the latent representation learned by deep neural networks with the KGs exploiting the semantic relationships between entities [  ... 
arXiv:2010.08660v4 fatcat:hcoahll2ivhdpcix7t6ezh425y

Bi-CLKT: Bi-Graph Contrastive Learning based Knowledge Tracing [article]

Xiangyu Song, Jianxin Li, Qi Lei, Wei Zhao, Yunliang Chen, Ajmal Mian
2022 arXiv   pre-print
With the recent rise of deep learning, Deep Knowledge Tracing (DKT) has utilised Recurrent Neural Networks (RNNs) to accomplish this task with some success.  ...  The goal of Knowledge Tracing (KT) is to estimate how well students have mastered a concept based on their historical learning of related exercises.  ...  Conventional approaches in this domain are mainly divided into the Bayesian Knowledge Tracing model using Hidden Markov Models [1] and Deep Knowledge Tracing using Deep Neural Networks [2] and its  ... 
arXiv:2201.09020v1 fatcat:se7pr4zmprh2hbokbgxdsgxriq

Depth-based signal separation with vertical line arrays in the deep ocean

Reid McCargar, Lisa M. Zurk
2013 Journal of the Acoustical Society of America  
Simulation results demonstrate depth-based signal separation without requiring knowledge of the ocean environment.  ...  Deep vertical line arrays can exploit the reliable acoustic path (RAP), which provides low transmission loss (TL) for targets at moderate ranges, and increased TL for distant interferers.  ...  No environmental knowledge was necessary, and the transform was computed assuming an isovelocity channel with c ¼ 1500 m/s.  ... 
doi:10.1121/1.4795241 pmid:23556698 fatcat:qwclbkq35jaubohm7v4oc4m6p4

TRACE: Early Detection of Chronic Kidney Disease Onset with Transformer-Enhanced Feature Embedding [article]

Yu Wang, Ziqiao Guan, Wei Hou, Fusheng Wang
2020 arXiv   pre-print
TRACE presents a comprehensive medical history representation with a novel key component: a Transformer-RNN autoencoder.  ...  In this paper, we propose the TRACE (Transformer-RNN Autoencoder-enhanced CKD Detector) framework, an end-to-end prediction model using patients' medical history data, to deal with these challenges.  ...  We summarize our contributions as follows: • To the best of our knowledge, this is the first work that builds advanced sequential deep learning models to predict CKD onset. • We propose a Transformer-RNN  ... 
arXiv:2012.03729v1 fatcat:hxohm7gepreg5pnxmj7yby4qk4

Integrate multi-omics data with biological interaction networks using Multi-view Factorization AutoEncoder (MAE)

Tianle Ma, Aidong Zhang
2019 BMC Genomics  
To alleviate the overfitting problem in deep learning on multi-omics data with the "big p, small n" problem, it is helpful to incorporate biological domain knowledge into the model as inductive biases.  ...  Our method learns feature and patient embeddings simultaneously with deep representation learning.  ...  Deep neural networks are often good at approximating any complex nonlinear transformations with appropriate training on a sufficiently large dataset.  ... 
doi:10.1186/s12864-019-6285-x pmid:31856727 pmcid:PMC6923820 fatcat:lmx32nlkwngvzgmv4tawn2yzei

Deep-IRT: Make Deep Learning Based Knowledge Tracing Explainable Using Item Response Theory [article]

Chun-Kit Yeung
2019 arXiv   pre-print
memory network (DKVMN) to make deep learning based knowledge tracing explainable.  ...  Deep learning based knowledge tracing model has been shown to outperform traditional knowledge tracing model without the need for human-engineered features, yet its parameters and representations have  ...  Deep Learning Based Knowledge Tracing Recently, with a surge of interest in deep learning models, deep knowledge tracing (DKT) [14] , which models student's knowledge state based on a recurrent neural  ... 
arXiv:1904.11738v1 fatcat:szgfp6ex45btjpogz7gawvqucm


Ruly Adha
2019 JL3T (Journal of Linguistics, Literature and Language Teaching)  
When those transformational rules are applied in a sentence, they will leave empty categories. Empty categories can be in the form of Complementizer (Comp), Trace, and PRO.  ...  Transformational rules consist of three types, namely movement transformation, deletion transformation, and substitution transformation.  ...  language(s); and the Theory of Language Use is concerned with the question of how linguistic and non-linguistic knowledge interact in speech comprehension and production.  ... 
doi:10.32505/jl3t.v5i1.887 fatcat:ejvgijxrmfedlhumrjkjvdfl4m

SAINT+: Integrating Temporal Features for EdNet Correctness Prediction [article]

Dongmin Shin, Yugeun Shim, Hangyeol Yu, Seewoo Lee, Byungsoo Kim, Youngduck Choi
2020 arXiv   pre-print
We propose SAINT+, a successor of SAINT which is a Transformer based knowledge tracing model that separately processes exercise information and student response information.  ...  Experimental results show that SAINT+ achieves state-of-the-art performance in knowledge tracing with an improvement of 1.25% in area under receiver operating characteristic curve compared to SAINT, the  ...  The advances of Deep Learning (DL) have given rise to neural network based knowledge tracing models. DKT [18] is the first DL based knowledge tracing model.  ... 
arXiv:2010.12042v1 fatcat:g22tl3cft5fo7pmdldcqf6g5ka
« Previous Showing results 1 — 15 out of 197,763 results