1,880 Hits in 7.4 sec

One Line To Rule Them All: Generating LO-Shot Soft-Label Prototypes [article]

Ilia Sucholutsky, Nam-Hwui Kim, Ryan P. Browne, Matthias Schonlau
2021 arXiv   pre-print
In addition, we propose the Hierarchical Soft-Label Prototype k-Nearest Neighbor classification algorithm based on these prototypical lines.  ...  Assigning soft labels to prototypes can allow increasingly small sets of prototypes to accurately represent the original training dataset.  ...  Thus, we propose the first method for generating prototypical lines and a new classification algorithm that can use them to perform LO-shot learning in practice.  ... 
arXiv:2102.07834v1 fatcat:nckkcy5z25fidi4hsa7krkgv34

'Less Than One'-Shot Learning: Learning N Classes From M<N Samples [article]

Ilia Sucholutsky, Matthias Schonlau
2020 arXiv   pre-print
We use a soft-label generalization of the k-Nearest Neighbors classifier to explore the intricate decision landscapes that can be created in the 'less than one'-shot learning setting.  ...  We propose the 'less than one'-shot learning task where models must learn N new classes given only M<N examples and we show that this is achievable with the help of soft labels.  ...  We believe that creating a soft-label prototype generation algorithm that specifically optimizes prototypes for LO-shot learning is an important next step in exploring this area.  ... 
arXiv:2009.08449v1 fatcat:24m2rrz6bbhrxihb4bl77t4fne

IJSD Volume 7 2017 All This: Full Issue

Claudia Kappenberg, Sarah Whatley, Harmony Bench, Simon Ellis
2016 The International Journal of Screendance  
</p><p>Please note: For the best offline reading experience, please print double-sided on 8.5x11 paper. Links to online sources are live in the PDF if read onscreen.</p></div>  ...  The rules to be followed by the pupils are not extensive at all: "One of the maxims of our school is: 'A little, but thoroughly,'" as Jakob reminds the reader, and: "To learn a little!  ...  A shot from the right side of the screen interacts with a shot from the left side of the screen. Movement is choreographed within those shots, but also for the relation they have to one another.  ... 
doi:10.18061/ijsd.v7i0.5480 fatcat:a4oxbqsibzf4zngp2jqwapy4fq

Syntactical and Semantical Description of Video Sequences [chapter]

N. Luth, A. Miene, P. Alshuth
1999 Database Semantics  
Our developed prototype system is planned to be integrated into the co-operative work of a TV-team of the TV-station "Radio Bremen".  ...  The latter is an useful approach for automatic trailer generation and for intelligent video editing as add-on for a video-cut-system.  ...  All succeeding shots will be compared to all previous shots of the clusters except the first previous one, because two consecutive shots are never taken from the same position or contain the same visual  ... 
doi:10.1007/978-0-387-35561-0_6 fatcat:lisfkwes6rfspbgxume3rjlzgu

Compositional Few-Shot Recognition with Primitive Discovery and Enhancing [article]

Yixiong Zou, Shanghang Zhang, Ke Chen, Yonghong Tian, Yaowei Wang, José M. F. Moura
2020 arXiv   pre-print
To modify this model for effective composition, inspired by both mathematical deduction and biological studies (the Hebbian Learning rule and the Winner-Take-All mechanism), we propose a soft composition  ...  Extensive experiments on public benchmarks are conducted on both the few-shot image classification and video recognition tasks.  ...  The model in each line on the left hand side consists of the modules all the above and the one listed in current line.  ... 
arXiv:2005.06047v3 fatcat:2k4nko7ztvboza2qyqekcmz66e

On the Ornithology of Central America. Part I

Philip Lutley Slater, Osbert Salvin
2008 Ibis  
In Europe between 30-40% of total IT budgets is spent on issues tied to Interoperability.  ...  Scope: The capability to efficiently interact, collaborate and exchange information with business partners and within a company is one of the most important challenges of each enterprise, especially forced  ...  Some of them are basic ones; others can be added by a domain expert. This flexibility permits its application to other domains. The prototype is based on multi-agents technology.  ... 
doi:10.1111/j.1474-919x.1859.tb06182.x fatcat:aedt43lqeffyzghkhhsjanjb5u

IRRODL Volume 13, Number 5

Various Authors
2012 International Review of Research in Open and Distance Learning  
Most likely in education, not all student works are subject to IP protection, but senior projects and master's and PhD theses could fall in to this category.  ...  Learners can also ask tutors and peers questions as well as discuss problems with them. Through sharing and communicating, learning information becomes large and complex.  ...  Acknowledgments We would like to thank three anonymous reviewers for their constructive comments and the editors for their time and valuable remarks.  ... 
doi:10.19173/irrodl.v13i5.1424 fatcat:lpe2ose6d5dlpea4xfhtxrbldi

Temporal video segmentation: A survey

Irena Koprinska, Sergio Carrato
2001 Signal processing. Image communication  
In addition to the algorithms for shot boundaries detection, the related topic of camera operation recognition is also reviewed.  ...  This article gives an overview of existing techniques for video segmentation that operate on both uncompressed and compressed video stream.  ...  Acknowledgements The "rst author was supported by a fellowship from the Consorzio per lo Sviluppo Internazionale dell'Universita`di Trieste.  ... 
doi:10.1016/s0923-5965(00)00011-4 fatcat:lnulrei7v5hybcufm7j3u4wyee

OptTyper: Probabilistic Type Inference by Optimising Logical and Natural Constraints [article]

Irene Vlassi Pandi, Earl T. Barr, Andrew D. Gordon, Charles Sutton
2021 arXiv   pre-print
To this end, we introduce a framework for probabilistic type inference that combines logic and learning: logical constraints on the types are extracted from the program, and deep learning is applied to  ...  We present a new approach to the type inference problem for dynamic languages.  ...  to the correct label from the output F and sum all the values of the correct labels for the entire training set.  ... 
arXiv:2004.00348v3 fatcat:jw3bkccforgaloehepwz6jfya4

"Once More unto the Breach, Dear Friends…": Olivier and Branagh's Henry V or Two Context-Dependent Renderings of the Shakespearean Original

Laura Torrado Mariñas
2017 ODISEA Revista de estudios ingleses  
Pointing out some of the various formal and structural alterations that Olivier imposed on his Henry V to make it a proper film for a post-war moment, and comparing them to Branagh's openly cinematic interpretation  ...  Para ello llamaremos la atención sobre las diversas estrategias, tanto formales como estructurales, que Olivier impone a su obra con la intención de adaptarla al contexto post-bélico de los años 40, y  ...  Following such a line of thought, the next scene I'd like to focus on is the attack on HarÀ eur.  ... 
doi:10.25115/odisea.v0i11.318 fatcat:lrgwp67otvgrrpccyt26k4ylcy

Finetuned Language Models Are Zero-Shot Learners [article]

Jason Wei, Maarten Bosma, Vincent Y. Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M. Dai, Quoc V. Le
2022 arXiv   pre-print
FLAN even outperforms few-shot GPT-3 by a large margin on ANLI, RTE, BoolQ, AI2-ARC, OpenbookQA, and StoryCloze.  ...  We show that instruction tuning -- finetuning language models on a collection of tasks described via instructions -- substantially improves zero-shot performance on unseen tasks.  ...  . • We moved prompt tuning from the main body to §4.5. • We added data contamination analysis (Appendix C). • We added few-shot instruction tuning ( §4.4). • We cited additional datasets in Appendix G.  ... 
arXiv:2109.01652v5 fatcat:lmnxjwtddnbbtnxqedb5uf57ye

Depicting as a method of communication

Herbert H. Clark
2016 Psychological review  
It is on a par with describing and pointing, but it works by different principles.  ...  To be complete, these models will have to account for depicting as well as describing and pointing.  ...  I wrote all these pieces in London, (raising pitch of voice) and I don't want any of you to play them on the pianos that you have here"?  ... 
doi:10.1037/rev0000026 pmid:26855255 fatcat:sqym7gcd7nbb5blljoqxrrkvly

Learning from Very Few Samples: A Survey [article]

Jiang Lu, Pinghua Gong, Jieping Ye, Changshui Zhang
2020 arXiv   pre-print
cognition to novelty from just a single or a handful of examples whereas machine learning algorithms typically entail hundreds or thousands of supervised samples to guarantee generalization ability.  ...  Finally, we conclude the survey with a discussion on promising trends in the hope of providing guidance and insights to follow-up researches.  ...  ACKNOWLEDGMENTS The authors would like to thank the pioneer researchers in few sample learning and other related fields.  ... 
arXiv:2009.02653v2 fatcat:fytfbeifmnhbfodat6czwxsqeu

An End-To-End Toolset For The Creation And Delivery Of Video-Based Multi-Device Content

I. Fraile
2018 Zenodo  
We introduce an end-to-end toolset for the production and delivery of synchronous multi-device content.  ...  Finally, we detail the system performance on different devices, and next steps.  ...  in this research in general.  ... 
doi:10.5281/zenodo.1402979 fatcat:gehlqemtongildlquzjz4dyzwe

The RQ-Tech methodology: a new paradigm for conceptualizing strategic enterprise architectures

Christine A. Hoyland, Kevin M. Adams, Andreas Tolk, Li D. Xu
2014 Journal of Management Analytics  
as one organic structure.  ...  A review of literature in all three areas provided an illustration of the overlap common to all three domains.  ...  Several screen shots of the RQ-Tech prototype are provided in Appendix E -RQ-Tech Prototype (Screen Shots).  ... 
doi:10.1080/23270012.2014.889912 fatcat:qj4yelfxnfeuzizfzdmsyanlki
« Previous Showing results 1 — 15 out of 1,880 results