11,311 Hits in 4.1 sec

Spelling approximate repeated or common motifs using a suffix tree [chapter]

Marie -France Sagot
1998 Lecture Notes in Computer Science  
Sketch of Extension Dealing with Gaps We sketch in this section how to extend the algorithms so as to be able to treat gaps as well as mismatches. This is done only for the repeated motifs problem.  ...  make such approaches prohibitive for big alphabets (if one dealt with proteins for instance instead of DNA sequences) and/or big values of k (as happens with some DNA signals such as the CRP binding site  ... 
doi:10.1007/bfb0054337 fatcat:zpy4notjqbcv7cgcuyekdndlyq

On the effect of model mismatch for sequential Info-Greedy Sensing

Ruiyang Song, Yao Xie, Sebastian Pokutta
2018 EURASIP Journal on Advances in Signal Processing  
Numerical examples demonstrate the good performance of Info-Greedy Sensing algorithms compared with random measurement schemes in the presence of model mismatch.  ...  We also show covariance sketching can be used as an efficient initialization for Info-Greedy Sensing.  ...  Availability of data and materials The Georgia Tech campus image is available atỹ xie77/campus.mat and the data for solar flare image is at www2.isye.gatech. edu/~yxie77/data_193.  ... 
doi:10.1186/s13634-018-0551-y fatcat:quitzpatkjehzhi2bq526a3foy

On Training Sketch Recognizers for New Domains [article]

Kemal Tugrul Yesilbek, T. Metin Sezgin
2021 arXiv   pre-print
Although transfer learning, and extensive data augmentation help deep learners, they still perform significantly worse compared to standard setups (e.g., SVMs and GBMs with standard feature representations  ...  Furthermore, we demonstrate that in realistic scenarios where data is scarce and expensive, standard measures taken for adapting deep learners to small datasets fall short of comparing favorably with alternatives  ...  Experiments show that training models with pseudo-realistic data fails to improve the recognizer performance when tested with realistic data compared to models trained with in-lab dataset.  ... 
arXiv:2104.08850v1 fatcat:yykeqpp6d5bdhhku4s32bf647q

On KDE-based Brushing in Scatterplots and how it Compares to CNN-based Brushing

Chaoran Fan, Helwig Hauser
2019 Workshop on Machine Learning Methods in Visualisation for Big Data  
In this paper, we investigate to which degree the human should be involved into the model design and how good the empirical model can be with more careful design.  ...  Based on this work, we then include a short discussion between the empirical model, designed in detail by an expert and the deep learning-based model that is learned from user data directly.  ...  We started with a large matrix of different combinations of the two parameters, covering a domain, which for sure was big enough.  ... 
doi:10.2312/mlvis.20191157 dblp:conf/mlvis-ws/FanH19 fatcat:yxozuordqngi5cpvpdupokwwlm

Sequential Information Guided Sensing [article]

Ruiyang Song, Yao Xie, Sebastian Pokutta
2015 arXiv   pre-print
Based on this, we further study how to estimate covariance based on direct samples or covariance sketching.  ...  Numerical examples also demonstrate the superior performance of Info-Greedy Sensing algorithms compared with their random and non-adaptive counterparts.  ...  Gaussian mixture model (GMM) We also establish a lower bound on the number of measurements (or power) required to recover a GMM signal with high precision when there is model mismatch.  ... 
arXiv:1509.00130v1 fatcat:siyx7lepkbhbda4xmrppy4dnki

Facial Synthesis from Visual Attributes via Sketch using Multi-Scale Generators [article]

Xing Di, Vishal M. Patel
2019 arXiv   pre-print
and (2) a face generator network which synthesizes facial images from the synthesized sketch images with the help of facial attributes.  ...  We first synthesize the facial sketch corresponding to the visual attributes and then we generate the face image based on the synthesized sketch.  ...  from the sketch generator at scale i, x i s ∼ P data (x i s ) stands for the real sketch image sampled from the sketch image data distribution at scale i,x i s is the attribute-mismatching sketch image  ... 
arXiv:1912.10479v1 fatcat:5hnmiv7qcvawrkifmgdusqy4ta

Big Data Analytics for Data Visualization: Review of Techniques

Geetika Chawla, Savita Bamal, Rekha Khatana
2018 International Journal of Computer Applications  
We can sketch a convincing visual story from raw data with the use of right tool. This paper focuses on Big Data visualization, its challenges, various tools.  ...  We have also investigated how virtual reality has radically changed the world of Big Data Visualization.  ...  Accumulating this large amount of data is very complex task. Unstructured data is another feature of big data. It is the information that does not have a pre-defined data model or format.  ... 
doi:10.5120/ijca2018917977 fatcat:adlcone3zzhuxnhdpzvnrefpl4

Specifying and Executing Application Behaviour with Condition-Request Rules

Andreas Harth, Tobias Käfer
2017 International Semantic Web Conference  
In projects we have routinely encountered obstacles for integration and interoperation due to architectural mismatches along several dimensions: network protocol, data format and data semantics.  ...  To specify applications, we present the syntax of a small language consisting of condition-request rules and sketch an operational semantics for the language based on an agent architecture with a sense-act  ...  Hence, the wrapper needs to lift the data model of the different components to a common data model. We assume the RDF data model.  ... 
dblp:conf/semweb/HarthK17 fatcat:574ipdejurdsrjs6mg3il4y2we

Pronunciation Proficiency Evaluation based on Discriminatively Refined Acoustic Models

Ke Yan, Shu Gong
2011 International Journal of Information Technology and Computer Science  
This paper introduces discriminative measures of minimum phone/word error (MPE/MWE) to refine acoustic models to deal with the problem.  ...  Experiments on the database of 498 people's live Putonghua test indicate that: 1) Refined acoustic models are more distinguishable than conventional MLE ones; 2) Even though training and test are mismatch  ...  Vividly speaking, it tells model "this is a big deep red apple, not a light red one". Fig.2 is a sketch map of the principal that how discriminative training improves scoring.  ... 
doi:10.5815/ijitcs.2011.02.03 fatcat:ikofgw5ryvbtbli3dtigm6h33q

Bind the gap: compiling real software to hardware FFT accelerators

Jackson Woodruff, Jordi Armengol-Estapé, Sam Ainsworth, Michael F. P. O'Boyle
2022 Proceedings of the 43rd ACM SIGPLAN International Conference on Programming Language Design and Implementation  
The fundamental issue is that of a mismatch between the diversity of user code and the functionality of fixed hardware, limiting its wider uptake.  ...  Speedups increase with data size as expected for an offloading-based accelerator model [19] .  ...  Significant mismatch also occurs at a datarepresentation level, data mismatch.  ... 
doi:10.1145/3519939.3523439 fatcat:6eoilci6knb3jczsx4dv577tia

Deep Sketch-Photo Face Recognition Assisted by Facial Attributes [article]

Seyed Mehdi Iranmanesh, Hadi Kazemi, Sobhan Soleymani, Ali Dabouei, Nasser M. Nasrabadi
2018 arXiv   pre-print
The results show the superiority of our method compared to the state- of-the-art models in sketch-photo recognition algorithms  ...  The proposed architecture is able to make full use of the sketch and complementary fa- cial attribute information to train a deep model compared to the conventional sketch-photo recognition methods.  ...  The deformation is performed by translating 25 preselected points with random magnitude and direction. 2-Scale and crop: One of the main mismatch problems between sketch images and their corresponding  ... 
arXiv:1808.00059v1 fatcat:aacadvlctbfrpkttkj2lgvt6qi

Smart, Deep Copy-Paste [article]

Tiziano Portenier and Qiyang Hu and Paolo Favaro and Matthias Zwicker
2019 arXiv   pre-print
Our training procedure works with any image dataset without additional information such as labels, and we demonstrate the effectiveness of our system on two popular datasets, high-resolution face images  ...  outperforms the current state of the art on face images, and we show promising results on the Cityscapes dataset, demonstrating that our system generalizes to much higher resolution than the training data  ...  Finally, we propose a training procedure in Section 3.3 that allows the training of our model end-to-end using suitable training data.  ... 
arXiv:1903.06763v1 fatcat:tvk4y2jvqjaxpi6523chiqj6oa

The Ouroboros Model, Selected Facets [chapter]

Knud Thomsen
2011 Advances in Experimental Medicine and Biology  
The Ouroboros Model features a biologically inspired cognitive architecture. At its core lies a self-referential recursive process with alternating phases of data acquisition and evaluation.  ...  Mismatches between anticipations based on previous experience and actual current data are highlighted and used for controlling the allocation of attention.  ...  Principal Algorithmic Backbone At the core of the Ouroboros Model lies a self-referential recursive process with alternating phases of data acquisition and evaluation.  ... 
doi:10.1007/978-1-4614-0164-3_19 pmid:21744223 fatcat:gm4lcbva7zbvfkgmpuu2nyu5hq

Cross-Modal Face Matching: Beyond Viewed Sketches [chapter]

Shuxin Ouyang, Timothy Hospedales, Yi-Zhe Song, Xueming Li
2015 Lecture Notes in Computer Science  
We approach this by learning a facial attribute model independently in each domain that represents faces in terms of semantic properties.  ...  Furthermore, we create a new dataset with ≈ 59, 000 attribute annotations for evaluation and to facilitate future research.  ...  flat-model performance (53 − 56% ).  ... 
doi:10.1007/978-3-319-16808-1_15 fatcat:dfwgwb2bujhcxocndjdq7k576a


2006 International journal of pattern recognition and artificial intelligence  
Some symbolic AI models for example BDI (belief, desire, intention) models are conceived as explicit and operational models of the intentional pursuit and belief dynamics.  ...  A clarification of the functional role of Surprise in a BDI-like cognitive architecture with respect to resource bounded belief revision is given.  ...  If the selected explanation ψ conflicts with the agent's expectation under scrutiny, the agent's expectation is invalidated due to the "logical" mismatch with the incoming interpretation of the input data  ... 
doi:10.1142/s0218001406004983 fatcat:6z5nzg4szfcnretbjcd52p632a
« Previous Showing results 1 — 15 out of 11,311 results