A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Filters
Interpreting and Boosting Dropout from a Game-Theoretic View
[article]
2021
arXiv
pre-print
Preserved Fulltext
Other Versions
Based on this understanding, we propose an interaction loss to further improve the utility of dropout. ...
This paper aims to understand and improve the utility of the dropout operation from the perspective of game-theoretic interactions. ...
Hinton et al. (2012) ; Srivastava et al. (2014) thought that dropout could encourage each unit in an intermediate-layer feature to model useful information without much dependence on other units. ...
arXiv:2009.11729v4
fatcat:wjr4rqe6ezec5fdsrzdjzmokwi
SI-VDNAS: Semi-Implicit Variational Dropout for Hierarchical One-shot Neural Architecture Search
2020
Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
In this paper, we propose a novel probabilistic approach, namely Semi-Implicit Variational Dropout one-shot Neural Architecture Search (SI-VDNAS), that leverages semi-implicit variational dropout to support ...
CSALR in Deep Metric Learning The proposed CSALR module can be easily applied to the existing backbones to capture long-range contextual dependencies efficiently, providing a general framework for deep ...
., 2017] , which can capture long-range contextual dependencies adaptively. ...
doi:10.24963/ijcai.2020/285
dblp:conf/ijcai/ChenGGD20
fatcat:3ca6iti3crbpxar246i5lkrcpy
Dropout of adult learners returning to university: interactions of motivational and environmental factors
2010
Bulletin of the Transilvania University of Braşov: Series VII: Social Sciences, Law
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The purpose of this paper is to highlight how motivational and contextual factors interact together to explain the dropout process of adult learners returning to university. ...
The findings from this study indicated that studying dropout of adult learners with motivational factors enables a deeper understanding taking into account the different commitments of this population ...
In the context of adult learners and according to the identity dynamic theory, their engagement depends on the relevance and efficiency of the educational program to reach their identity goals. ...
doaj:770db3d6385d44ada4c72f825f86f729
fatcat:eetg4oigoravxirqvx7zogypja
Contextual RNN-T For Open Domain ASR
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
We evaluate our approach on an in-house dataset sampled from de-identified public social media videos, which represent an open domain ASR task. ...
By using an attention model and a biasing model to leverage the contextual metadata that accompanies a video, we observe a relative improvement of about 16% in Word Error Rate on Named Entities (WER-NE ...
(EE), an Attention Module (AttModule) and a Biasing Module (BiasingModule) as shown in Figure 2 . ...
arXiv:2006.03411v2
fatcat:cwjn4eeiyrc4lcraeeqx6bg2qa
Maps Search Misspelling Detection Leveraging Domain-Augmented Contextual Representations
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Building an independent misspelling detector and serve it before correction can bring multiple benefits to speller and other search components, which is particularly true for the most commonly deployed ...
With rapid development of deep learning and substantial advancement in contextual representation learning such as BERTology, building a decent misspelling detector without having to rely on hand-crafted ...
Algorithm The central mission for such an algorithm is to generate high quality pairs, Q(misspelt query) -> C(ground-truth correction), freely, accurately and at large scale, no dependency on human judgement ...
arXiv:2108.06842v1
fatcat:4crfyqzlvbhirebragmeoekgbu
Contextual RNN-T for Open Domain ASR
2020
Interspeech 2020
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We evaluate our approach on an in-house dataset sampled from deidentified public social media videos, which represent an open domain ASR task. ...
By using an attention model to leverage the contextual metadata that accompanies a video, we observe a relative improvement of about 16% in Word Error Rate on Named Entities (WER-NE) for videos with related ...
Contextual RNN-T We modify the base RNN-T model described in Section 3 and add three additional components: an Embedding Extractor (EE), an Attention Module (AttModule) and (optionally) a Biasing Module ...
doi:10.21437/interspeech.2020-2986
dblp:conf/interspeech/JainKMZMS20
fatcat:pluyzzp47ngfhgh4ws4277ofzy
Dissecting Contextual Word Embeddings: Architecture and Representation
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Contextual word representations derived from pre-trained bidirectional language models (biLMs) have recently been shown to provide significant improvements to the state of the art for a wide range of NLP ...
We show there is a tradeoff between speed and accuracy, but all architectures learn high quality contextual representations that outperform word embeddings for four challenging NLP tasks. ...
Billion Word Benchmark (Chelba et al., 2014) using a sampled softmax with 8192 negative samples per batch. ...
arXiv:1808.08949v2
fatcat:ogd5bpuwmjfd7j57lnrq6fxfxq
Time-Series Representation Learning via Temporal and Contextual Contrasting
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Last, to further learn discriminative representations, we propose a contextual contrasting module built upon the contexts from the temporal contrasting module. ...
In this paper, we propose an unsupervised Time-Series representation learning framework via Temporal and Contextual Contrasting (TS-TCC), to learn time-series representation from unlabeled data. ...
In this contextual contrasting module, we aim to maximize the similarity among different contexts of the same sample while minimizing similarity among contexts of different samples. ...
arXiv:2106.14112v1
fatcat:b45ts7bwungevd5jn375jnbmba
Exploiting Contextual Information with Deep Neural Networks
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
To the best of our knowledge, we are the first to integrate graph-theoretical modules, carefully crafted for the problem of similarity learning and that are designed to consider contextual information, ...
Nevertheless, there has not been much research in exploiting contextual information in deep neural networks. ...
Another widely used deep learning-specific form of regularization is dropout [173] . Dropout in forward pass simply drops units with probability p, making every unit less dependent in its neighbors. ...
arXiv:2006.11706v2
fatcat:bzaghxubbzcftdsvv4ogk3tfxe
What Predicts Clinician Dropout from State-Sponsored Managing and Adapting Practice Training
2015
Administration and Policy in Mental Health
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Dropouts from system-wide evidence-based practice trainings are high; yet there are few studies on what predicts dropouts. ...
In 2006, NYS Office of Mental Health (OMH) established an evidence-based treatment (EBT) training center for mental health providers statewide. ...
Clinicians' affiliation was also significantly different depending on region (Fishers exact <.001). ...
doi:10.1007/s10488-015-0709-y
pmid:26699136
pmcid:PMC5545802
fatcat:f3vvpbvcnvdfvg77ybg3aphu6a
Patterns, Consequences, and Possible Causes of Dropout in Upper Secondary Education in Mexico
2015
Education Research International
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
The present study provides a detailed analysis of upper secondary education dropout patterns in Mexico, exploring its consequences and possible causes. ...
The main value added is the simultaneous analysis of the influence of individual-family, community, and macroaggregate factors, on school dropout in the country. ...
Acknowledgments The authors thank the helpful comments and suggestions by an anonymous referee, which helped improve the paper substantially. ...
doi:10.1155/2015/676472
fatcat:ktmisnorlbhrbmc7chf22tky4m
Understanding in-video dropouts and interaction peaks inonline lecture videos
2014
Proceedings of the first ACM conference on Learning @ scale conference - L@S '14
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2014; you can also visit the original URL.
The file type is application/pdf.
We find higher dropout rates in longer videos, re-watching sessions (vs first-time), and tutorials (vs lectures). ...
In attempting to reason why peaks occur by sampling 80 videos, we observe that 61% of the peaks accompany visual transitions in the video, e.g., a slide view to a classroom view. ...
While our categorization is not conclusive, it provides an explanation of which semantic and contextual aspects of video might be responsible for a peak. ...
doi:10.1145/2556325.2566237
dblp:conf/lats/KimGSMGM14
fatcat:ys2zi3my65he5gjq6qzsufipgy
Pragmatic inference and visual abstraction enable contextual flexibility during visual communication
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Together, this work provides the first algorithmically explicit theory of how visual perception and social cognition jointly support contextual flexibility in visual communication. ...
We found that people exploited shared information to efficiently communicate about the target object: on far trials, sketchers achieved high recognition accuracy while applying fewer strokes, using less ...
We use MCMC to draw 1000 samples from the joint posterior with a lag of 0, discarding 3000 burn-in samples. ...
arXiv:1903.04448v2
fatcat:dk2npj6j6jcb7clhgesjv73vdu
Sato: Contextual Semantic Type Detection in Tables
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
However, existing detection approaches either perform poorly with dirty data, support only a limited number of semantic types, fail to incorporate the table context of columns or rely on large sample sizes ...
Additionally, we believe that high-order dependency between predictions is not always necessary if we incorporate contextual features into the model. ...
Since table topic features provide table-wise contextual information, we consider the original CRF model with pairwise potential functions as the right choice for improving the model accuracy efficiently ...
arXiv:1911.06311v3
fatcat:q763z33jnvbyhjqojpuqhl3uge
Developing a model for dropout prevention and intervention in primary and secondary schools in Serbia: Assessing the model's effectiveness
2017
Psihološka Istraživanja
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Understanding dropout as a multidimensional and system-level phenomenon, we offer a model for dropout prevention and intervention. ...
The paper focuses on investigating the dropout phenomena and the need for its prevention and reduction within Serbian pre-university education. ...
students at dropout risk, drafts an individual plan for dropout prevention (IPDP). ...
doi:10.5937/psistra1701145m
fatcat:cae2yo7pgvg7lpniem52atnemy
« Previous
Showing results 1 — 15 out of 4,625 results