Filters








11,396 Hits in 6.3 sec

Learn to Focus: Hierarchical Dynamic Copy Network for Dialogue State Tracking [article]

Linhao Zhang, Houfeng Wang
2021 arXiv   pre-print
In this paper, we propose a Hierarchical Dynamic Copy Network (HDCN) to facilitate focusing on the most informative turn, making it easier to extract slot values from the dialogue context.  ...  Recently, researchers have explored using the encoder-decoder framework to tackle dialogue state tracking (DST), which is a key component of task-oriented dialogue systems.  ...  To this end, we propose a Hierarchical Dynamic Copy Network (HDCN) to facilitate focusing on the most informative turns, making it easier to copy slot values from the dialogue context.  ... 
arXiv:2107.11778v1 fatcat:5wxiswur7beunhfum4okukrrpm

Learning to Memorize in Neural Task-Oriented Dialogue Systems [article]

Chien-Sheng Wu
2019 arXiv   pre-print
In this thesis, we leverage the neural copy mechanism and memory-augmented neural networks (MANNs) to address existing challenge of neural task-oriented dialogue learning.  ...  We also evaluate unseen domain dialogue state tracking and show that TRADE enables zero-shot dialogue state tracking and can adapt to new few-shot domains without forgetting the previous domains.  ...  dynamic query memory network, with recorded delexicalization copying for end-to-end retrievalbased dialogue learning.  ... 
arXiv:1905.07687v1 fatcat:upvqqczyajdtnf45bcszxs6nai

Taming the Beast: Learning to Control Neural Conversational Models [article]

Andrea Madotto
2021 arXiv   pre-print
, natural language understanding (NLU), dialogue state tracking (DST), a dialogue manager (DM) and natural language generation (NLG).  ...  Therefore, in this thesis, we study controllable methods for end-to-end generative dialogue systems in task-oriented and chit-chat scenarios.  ...  recognition, state tracking, natural language generation, and end-to-end.  ... 
arXiv:2108.10561v1 fatcat:ixton6i3gbhv7kri4pgaylyybi

Learning to Describe Solutions for Bug Reports Based on Developer Discussions [article]

Sheena Panthaplackel, Junyi Jessy Li, Milos Gligoric, Raymond J. Mooney
2022 arXiv   pre-print
With automated and human evaluation, we find this task to form an ideal testbed for complex reasoning in long, bimodal dialogue context.  ...  We build a corpus for this task using a novel technique for obtaining noisy supervision from repository changes linked to bug reports, with which we establish benchmarks.  ...  Hier S2S + Ptr: Inspired by hierarchical approaches for dialogue response generation (Serban et al. 2016 ), we consider a hierarchical variant of the S2S + Ptr model with two separate encoders: one that  ... 
arXiv:2110.04353v2 fatcat:mesfaxszpbbc7by453kvg2uude

Machine learning techniques to make computers easier to use

Hiroshi Motoda, Kenichi Yoshida
1998 Artificial Intelligence  
The last two sections consider lessons learned from this study and directions for future research.  ...  Prefetch duemon works as expected only for I/O intensive task due to an implementation problem, and thus needs further improvement. The following section introduces the three learning tasks.  ...  Authors are grateful to Shojiro Asai and Katsuki Miyauchi of ARL for their generous support of this work.  ... 
doi:10.1016/s0004-3702(98)00062-9 fatcat:3egioabdfjbvlpyjlmxazmdoce

FOCUS ON THE ESSENTIALS

James W. Cortada
1998 Journal of Business Strategy  
FOCUS ON THE ESSENTIALS James W. Cortada You wont have to turn yourself into a propeller-head, but it would behoove you to sharpen your focus on technology in the coming year.  ...  Cortada, Ph.1 We can already see the first fruits of e-commerce in Federal Express’s use of networks and wireless commu- nications to track packages; in Ben ‘n Jerry's manufacture of ice cream fla- vors  ... 
doi:10.1108/eb039971 fatcat:crn7wvzvmbbbhl6f67fxyg6yq4

A Gentle Introduction to Reinforcement Learning and its Application in Different Fields

Muddasar Naeem, S. Tahir H. Rizvi, Antonio Coronato
2020 IEEE Access  
For example, in DQN the Q-estimates are computed using an old copy of the network (a "target" network), that will provide "older" Qestimates, with a very specific kind of bias, relating to the belief of  ...  Main focus of this algorithm was to deal with the different environmental variables having large state space. Variable estimation is the main goal to be addressed.  ... 
doi:10.1109/access.2020.3038605 fatcat:febm7kz525adpcvkfmnim2yha4

Explicit State Tracking with Semi-Supervisionfor Neural Dialogue Generation

Xisen Jin, Wenqiang Lei, Zhaochun Ren, Hongshen Chen, Shangsong Liang, Yihong Zhao, Dawei Yin
2018 Proceedings of the 27th ACM International Conference on Information and Knowledge Management - CIKM '18  
dialogues, most of existing work neglects the explicit state tracking due to the unlimited number of dialogue states.  ...  Tracking dialogue states is an important ingredient in dialogue generation for estimating users' intention.  ...  Thus in neural dialogue generation, dynamically tracking dialogue states is the key for generating coherent and context-sensitive responses.  ... 
doi:10.1145/3269206.3271683 dblp:conf/cikm/JinLRCLZY18 fatcat:reztbc7xefhqtpfe6ixqm6ui4m

Innovative Pathways to the Next Level of e-Learning

Victoria Nash, William H. Dutton, Malcolm Peltu
2004 Social Science Research Network  
This paper aims to clarify the key issues of that debate, which started with computer-aided instruction systems in the 1960s, in order to offer a better understanding of the 'next level' of e-learning  ...  , universities, libraries, households and other locations where e-learning can take place.  ...  for Standards in Education (UK) P2P Peer-to-Peer (networks), as opposed to networks with more centralized and hierarchical architectures SAT Standard Assessment/Scholastic Aptitude Test Smartboard Alternative  ... 
doi:10.2139/ssrn.1325224 fatcat:th5xizntnnbalnlhm6243sfhxm

Using activity-oriented design methods to study collaborative knowledge-building in e-learning courses within higher education

Christine Greenhow, Brad Belbas
2007 International Journal of Computer-Supported Collaborative Learning  
As researchers and educational technologists working in a large public research university, our research focuses on understanding the interactions among various stakeholder groups involved in e-Learning  ...  Trends in higher education have contributed to the need for more coordination and collaboration among different constituencies involved in instructional design and delivery.  ...  networks (Bernard et al. 2004 ).  ... 
doi:10.1007/s11412-007-9023-3 fatcat:ywhgqfvrcvdxrbfb7pajjyxuyy

Translational Approach to Behavioral Learning: Lessons from Cerebellar Plasticity

Guy Cheron, Bernard Dan, Javier Márquez-Ruiz
2013 Neural Plasticity  
deciphering fundamental and translational mechanisms from cellular to network levels.  ...  The privileged relationship between the cerebellum and the inferior olive offers an ideal circuit for attempting to integrate the numerous evidences of neuronal plasticity into a translational perspective  ...  In line with hierarchical network organizations, Huang et al. [113] propose that this GC sensory-motor integration may allow the PC to accomplish more complex learning tasks.  ... 
doi:10.1155/2013/853654 pmid:24319600 pmcid:PMC3844268 fatcat:lgtoa7glijagxprunhhlb4opwu

The Dynamics of Group Learning in a Cohort: From Nonlearning to Transformative Learning

Jay Paredes Scribner, Joe F. Donaldson
2001 Educational Administration Quarterly  
The overall goal is to promote sound scholarship and clarity of expression in an open and continuing dialogue among scholars and practitioners in all kinds of educational organizations.  ...  reform, governance and reform in colleges of education, the teaching of educational administration, and the professional preparation of educational administrators-topics advancing the University Council for  ...  The items asked the respondents to estimate the extent to which their jobs provided opportunities for (a) learning new jobs, (b) working in the way they thought best, (c) using their abilities, (d) using  ... 
doi:10.1177/00131610121969442 fatcat:nmbbszjxezabrb33cw4ltqqf7a

A Progressive Digital Narrative Teaching Method to Improve Learning Motivation as a Lifelong Learning Skill

Chih-Hao Lin, Yao-Yun Chang
2021 Sustainability  
The objective of this research is to adopt a step-by-step three-stage teaching method to gradually increase students' interest in learning and make computer network learning more effective.  ...  Traditional teaching methods are not suitable for business school students who think the content is too theoretical to understand.  ...  When faced with the rapid response needs and active learning characteristics of Generation Z students, it is difficult for traditional teaching methods to adjust dynamically in response to students' learning  ... 
doi:10.3390/su132312991 fatcat:ejgaz5zrbvcrjkvj6vkn3364tq

Connect to Learn: Assemblage of Pedagogies in Higher Education in a Community of Practice

Elaine Fournier, Mina Sedaghatjou, Immaculate Namukasa
2021 The Open/Technology in Education, Society, and Scholarship Association Journal  
The goal of the Community of Practice is to connect with and learn from one another, discussing, and reflecting on different types of pedagogical practices among members who teach in both graduate and  ...  In this paper, we report on the ways in which technology and scholarship of pedagogy emerge as interconnected within a technology-facilitated community of practice (CoP), for educators within various Faculties  ...  COPM1 made a poignant declaration when she stated: Participating in ongoing collaboration, dialogue, mutual trust, social networking, helping, and learning from colleagues, all of those  ... 
doi:10.18357/otessaj.2021.1.2.10 fatcat:fuu7grjybbb3rb37o3bj6ulsue

Using Social Network Analysis to Explore Digital Student Interactions and Business Competency Learning in a Web-based Educational Platform

Andrew Feldstein, Kim Gower
2015 International Journal of Information Systems and Social Change  
The resulting .csv files (execrted in Table 1 ) were read by Gephi, which is "an interactive visualization and exploration platform for all kinds of networks and complex systems, dynamic and hierarchical  ...  As stated earlier, the goal of this study was to map the comment network in a specific social learning community as a first step towards developing a more versatile range of assessment tools.  ... 
doi:10.4018/ijissc.2015010101 fatcat:ohnj3eunfnhnrf5pp66i6rfl5a
« Previous Showing results 1 — 15 out of 11,396 results