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.
Filters
Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification
[article]
2020
arXiv
pre-print
Preserved Fulltext
Other Versions
State-of-the-art unsupervised domain adaptation methods for person re-ID transferred the learned knowledge from the source domain by optimizing with pseudo labels created by clustering algorithms on the ...
better features from the target domain via off-line refined hard pseudo labels and on-line refined soft pseudo labels in an alternative training manner. ...
Teacher-student models have been widely studied in semi-supervised learning methods and knowledge/model distillation methods. ...
arXiv:2001.01526v2
fatcat:7nqggtsmwvhp3kkz5ofm76xuli
The Knowledge Based Agricultural Bioeconomy: A Bibliometric Network Analysis
2021
Energies
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.
The label 'knowledge-based', not so widely used at EU level, is explicitly used in some national strategies, and echoed in others, as all strategies focus on new scientific knowledge and technologies and ...
Discussion In 2015-2016, research on bioeconomy focused on bioenergy and biofuels, since the primary objective was to find new sources of renewable fuels and to decouple from fossil fuels, the quantities ...
doi:10.3390/en14206823
fatcat:k3u4v4jxsbbmri4dcnotf4qb5e
Rethinking Re-Sampling in Imbalanced Semi-Supervised Learning
[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.
Other Versions
Semi-Supervised Learning (SSL) has shown its strong ability in utilizing unlabeled data when labeled data is scarce. ...
We find that data re-sampling is of critical importance to learn a good classifier as it increases the accuracy of the pseudo-labels, in particular for the minority classes in the unlabeled data. ...
aims at transferring knowledge from head classes to tail classes. ...
arXiv:2106.00209v2
fatcat:gibpzulpbveebdvnbmrp6ivpza
CoSSL: Co-Learning of Representation and Classifier for Imbalanced Semi-Supervised Learning
[article]
2022
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 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In this paper, we propose a novel co-learning framework (CoSSL) with decoupled representation learning and classifier learning for imbalanced SSL. ...
In experiments, we show that our approach outperforms other methods over a large range of shifted distributions, achieving state-of-the-art performance on benchmark datasets ranging from CIFAR-10, CIFAR ...
[30] proposed Distribution Aligning Refinery (DARP) to refine raw pseudo-labels via convex optimization. Wei et al. ...
arXiv:2112.04564v3
fatcat:cyfqgnld75hyrpcj5q2siptcca
Investing in the Future of Jobs and Skills; Scenarios, Implications and Options in Anticipation of Future Skills and Knowledge Needs - Sector Report Chemicals, Pharmaceuticals, Rubber & Plastic Products
2010
Social Science Research Network
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.
requirements (sometimes labelled as 'hard skills'). ...
This "decoupling" of the chemicals and pharmaceuticals industry is the result of a number of factors. ...
Knowledge. Knowledge refers to the outcome of the accumulation of information through learning. ...
doi:10.2139/ssrn.1579302
fatcat:mob53mxtpfbcrcgss3tsvt4h24
Knowledge from Data, Knowledge from Doing: The Inclusionary Production of Environmental Knowledge for Management
2017
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.
Decoupling the physical labor of practice from its knowledge-making dimensions may exacerbate the problem of transience in volunteers. ...
Etienne Wenger (1998) , along with Jean Lave (Lave and Wenger 1991) , proposed the label "communities of practice" to describe a holistic view of learning and knowledge production as emerging from the ...
doi:10.7298/x4v69gqq
fatcat:n2kfyj6vdbfdhnvorofqti6mqq
Multi-organ Segmentation via Co-training Weight-averaged Models from Few-organ Datasets
[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.
In this paper, we propose to co-train weight-averaged models for learning a unified multi-organ segmentation network from few-organ datasets. ...
To alleviate the noisy teaching supervisions between the networks, the weighted-averaged models are adopted to produce more reliable soft labels. ...
The key idea is to transfer knowledge from a teacher to a student network via consistency training. ...
arXiv:2008.07149v1
fatcat:ghrbzwhgm5emhbtahqrf76zb2e
Towards Practical Lipreading with Distilled and Efficient Models
[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.
Other Versions
Thirdly, we show that knowledge distillation is a very effective tool for recovering performance of the lightweight models. ...
“Label refinery: Improving ImageNet classification through
[3] S. Petridis, Y. Wang, Z. Li, and M. Pantic, “End-to-end multi- label progression,” arXiv:1805.02641, 2018. ...
to transfer knowledge from a teacher
LRW1000 datasets. ...
arXiv:2007.06504v3
fatcat:7xixuerqorbxlhimvmw3nlcupe
The pathway to a sustainable industrial society – initiative of the Research Institute for Sustainability Science (RISS) at Osaka University
2006
Sustainability Science
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.
Combinations of existing knowledge account for most innovation as a result of various forms of learning: learning-by-doing (Arrow 1962) , learning-by-using (Rosenberg 1982) and learning-by-interacting ...
The scenario studies seek to ensure sustainable development by decoupling economic development from environmental pressure. 2. ...
doi:10.1007/s11625-006-0008-3
fatcat:ts2ck3rflje7linzambg3lj3vu
Differential Impact of Environmental Policy Instruments on Technological Change: A Review of the Empirical Literature
2007
Social Science Research Network
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 survey is explicitly focused on the empirical identification of the hypothesis to expect a stronger impact from market-based incentives than from non-market alternatives. ...
The general picture emerging from the recent literature is that there is a clear impact of environmental policy on invention, innovation and diffusion of technologies. ...
The second mechanism is labelled 'learning by doing', but this concept is somewhat more general here, different countries. ...
doi:10.2139/ssrn.991612
fatcat:j5uqqtfjofgprfgpi6535oyqhu
CReST: A Class-Rebalancing Self-Training Framework for Imbalanced Semi-Supervised Learning
[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.
Other Versions
CReST iteratively retrains a baseline SSL model with a labeled set expanded by adding pseudo-labeled samples from an unlabeled set, where pseudo-labeled samples from minority classes are selected more ...
While existing semi-supervised learning (SSL) methods are known to perform poorly on minority classes, we find that they still generate high precision pseudo-labels on minority classes. ...
Some works [50, 44, 23, 28, 29] aim to transfer knowledge from majority classes to minority classes. ...
arXiv:2102.09559v2
fatcat:m6hf33cgcrcavjtcc6crszbhvy
Emission Quantification via Passive Infrared Optical Gas Imaging: A Review
2022
Energies
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 2022; you can also visit the original URL.
The file type is application/pdf.
Advanced AI algorithms, as well as the simultaneous development of diagnostics based on relevant physics and AI algorithms for the accurate and correct extraction of quantitative information from infrared ...
intelligence (AI) with quantitative IOGI in all aspects, which substantially enhances the feasibility, performance, and agility of quantitative IOGI, and alleviates its heavy reliance on prior context-based knowledge ...
First, from the pixel value of an IR image, we cannot decouple the concentration and light path length-they are represented by column density. ...
doi:10.3390/en15093304
fatcat:xkt5gbp2rzelpkj7xevrrytn3e
Circular Economy across Australia: Taking Stock of Progress and Lessons
2021
Circular Economy and Sustainability
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.
across the supply chain, technological constraints, lack of policy coordination for business innovation, economic recession induced by the COVID-19 pandemic and lack of incentives to change behavior from ...
Figure 4 shows the heavy industries involved, which include an alumina refinery (Alcoa), a nickel refinery (Kwinana Nickel Refinery), a titanium dioxide pigment plant (Tiwest), lime and cement kilns ( ...
Cockburn Cement), an oil refinery (BP) and a pig iron plant (HIsmelt). ...
doi:10.1007/s43615-021-00020-5
pmid:34888553
pmcid:PMC7968550
fatcat:pogtxpkimne2tgisfrw7nqwwau
From Smoke and Mirrors to Walking the Talk: Decoupling in the Contemporary World
2012
The Academy of Management Annals
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.
What can we learn from this active strand of research on decoupling and the relationship between policy and practice? ...
Snellman (2011) labels this "reverse decoupling" and illuminates it in a study of the embrace of shareholder value practices in Finnish corporate governance. ...
doi:10.1080/19416520.2012.684462
fatcat:5u3fd4fozrbn7amjd2slvlcjoe
From Smoke and Mirrors to Walking the Talk: Decoupling in the Contemporary World
2012
The Academy of Management Annals
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.
What can we learn from this active strand of research on decoupling and the relationship between policy and practice? ...
Snellman (2011) labels this "reverse decoupling" and illuminates it in a study of the embrace of shareholder value practices in Finnish corporate governance. ...
doi:10.5465/19416520.2012.684462
fatcat:7laicvb43vcbhf4s7vwxxr3da4
« Previous
Showing results 1 — 15 out of 388 results