Filters








771,855 Hits in 6.8 sec

Strong separation of learning classes [chapter]

John Case, Keh-Jiann Chen, Sanjay Jain
1992 Lecture Notes in Computer Science  
A philosophical heuristic toward the design of artificially intelligent learning programs is presented with each strong separation result.  ...  It is shown that for many such separated learning classes from the literature a much stronger separation holds: (∀C ∈ It is also shown that there is a pair of separated learning classes from the literature  ...  Fortunately, in the present paper, for a wide variety of learning classes LC 1 , LC 2 such that LC 1 ⊆ LC 2 and (LC 2 − LC 1 ) = ∅, we show the strong separation result that (∀C ∈ LC 1 )(∃C ∈ (LC 2 − LC  ... 
doi:10.1007/3-540-56004-1_9 fatcat:pirzqs5jlvgyjk6ttmgomclpii

Strong separation of learning classes

JOHN CASE, KEH-JIANN CHEN, SANJAY JAIN
1992 Journal of experimental and theoretical artificial intelligence (Print)  
A philosophical heuristic toward the design of artificially intelligent learning programs is presented with each strong separation result.  ...  It is shown that for many such separated learning classes from the literature a much stronger separation holds: (∀C ∈ It is also shown that there is a pair of separated learning classes from the literature  ...  Fortunately, in the present paper, for a wide variety of learning classes LC 1 , LC 2 such that LC 1 ⊆ LC 2 and (LC 2 − LC 1 ) = ∅, we show the strong separation result that (∀C ∈ LC 1 )(∃C ∈ (LC 2 − LC  ... 
doi:10.1080/09528139208953752 fatcat:r5k4rl62brg5rmohtsssbolxuq

Proactive Password Strength Analyzer Using Filters and Machine Learning Techniques

G Suganya, S Karpgavalli, V Christina
2010 International Journal of Computer Applications  
This framework can be implemented as a submodule of the access control scheme. General Terms Password strength, machine learning techniques  ...  Thus it is important for organizations to recognize the vulnerabilities to which passwords are subjected, and develop strong policies governing the creation and use of passwords to ensure that those vulnerabilities  ...  SVMs construct a hyperplane that separates two classes and tries to achieve maximum separation between the classes.  ... 
doi:10.5120/1333-1788 fatcat:lwijfuybhnhtrihg4tr4ngaas4

Weight-Based Facial Expression Recognition from Near-Infrared Video Sequences [chapter]

Matti Taini, Guoying Zhao, Matti Pietikäinen
2009 Lecture Notes in Computer Science  
The use of LBP-TOP features enables us to set different weights for each of the three planes (appearance, horizontal motion and vertical motion) inside the block volume.  ...  Facial expressions can be thought of as specific dynamic textures where local appearance and motion information need to be considered.  ...  The financial support provided by the European Regional Development Fund, the Finnish Funding Agency for Technology and Innovation and the Academy of Finland is gratefully acknowledged.  ... 
doi:10.1007/978-3-642-02230-2_25 fatcat:nplhteldgzgr3m43rlavt7x3wq

Decoupled Deep Neural Network for Semi-supervised Semantic Segmentation [article]

Seunghoon Hong, Hyeonwoo Noh, Bohyung Han
2015 arXiv   pre-print
Contrary to existing approaches posing semantic segmentation as a single task of region-based classification, our algorithm decouples classification and segmentation, and learns a separate network for  ...  The decoupled architecture enables us to learn classification and segmentation networks separately based on the training data with image-level and pixel-wise class labels, respectively.  ...  In the standard semi-supervised learning framework, given only a small number of training images with strong annotations, one needs to infer the full segmentation labels for the rest of the data.  ... 
arXiv:1506.04924v2 fatcat:kdljwoets5h53jes3iwph75ypy

Improving the separability of a reservoir facilitates learning transfer

David Norton, Dan Ventura
2009 2009 International Joint Conference on Neural Networks  
Instead, the liquid is evaluated based on its ability to separate different classes of input into different spatial patterns of neural activity.  ...  We use a type of reservoir computing called the liquid state machine (LSM) to explore learning transfer.  ...  delay, the number of samples per class used to determine separation at each instance, the number of iterations to run, the learning rate, the decay time constant, and the amount of noise present in each  ... 
doi:10.1109/ijcnn.2009.5178656 dblp:conf/ijcnn/NortonV09 fatcat:6mgszzdybjetricyo7mosegvwy

Limit Learning Equivalence Structures [article]

Ekaterina Fokina, Timo Kötzing, Luca San Mauro
2019 arXiv   pre-print
Finally, we show how learning families of structures relates to learning classes of languages by mapping learning tasks for structures to equivalent learning tasks for languages.  ...  While most research in Gold-style learning focuses on learning formal languages, we consider the identification of computable structures, specifically equivalence structures.  ...  of structures with infinite equivalence classes, we get a separation of learning power. ( 1) A is Inf Ex ∼ -learnable. (2) A is TxtEx ∼ -learnable.  ... 
arXiv:1902.08006v1 fatcat:rtcksf3qmnfjrpbihpxtbo5mmy

Latent Representations of Terrain in Aerial Image Classification

Pylyp Prystavka, Serge Dolgikh, Olga Cholyshkina, Oleksandr Kozachuk
2021 International Conference on Information and Communication Technologies in Education, Research, and Industrial Applications  
classes of terrain types were studied.  ...  The analysis of distributions demonstrated a landscape of compact concept clusters for most studied types of terrain with good separation between concept regions.  ...  generative learning and provide strong additional arguments in support of the general character of this effect.  ... 
dblp:conf/icteri/PrystavkaDCK21 fatcat:x6hupwx5bzh5hlzlfrtmy5wfvu

Finding Strength in Weakness: Learning to Separate Sounds with Weak Supervision [article]

Fatemeh Pishdadian, Gordon Wichern, Jonathan Le Roux
2019 arXiv   pre-print
While there has been much recent progress using deep learning techniques to separate speech and music audio signals, these systems typically require large collections of isolated sources during the training  ...  Our objective function requires the classifier applied to a separated source to assign high probability to the class corresponding to that source and low probability to all other classes.  ...  separator to output estimated sources that match the previously learned representation for each sound class.  ... 
arXiv:1911.02182v1 fatcat:247o2azm5zerrkba7miu7umcua

Increasing the power of uniform inductive learners

Sandra Zilles
2005 Journal of computer and system sciences (Print)  
Modifications of the formal constraints affect the quality of the corresponding learners on the one hand and regulate the quantity of learnable classes on the other hand.  ...  For technical reasons these results are achieved using various formal conditions concerning the learnability of unions of uniformly learnable classes.  ...  Thomas Zeugmann, as well as the anonymous referees for many useful hints contributing to the quality of the presentation.  ... 
doi:10.1016/j.jcss.2004.10.004 fatcat:o2sp554fl5fclag2nl6tl3s2mu

Page 4768 of Mathematical Reviews Vol. , Issue 94h [page]

1994 Mathematical Reviews  
of learning classes.  ...  It is also shown that there is a pair of separated learning classes from the literature for which the stronger separation above does not hold.  ... 

Ordinal Classification Method for the Evaluation Of Thai Non-life Insurance Companies [article]

Phaiboon Jhonpita, Sukree Sinthupinyo, Thitivadee Chaiyawat
2012 arXiv   pre-print
This paper proposes a use of an ordinal classifier to evaluate the financial solidity of non-life insurance companies as strong, moderate, weak, and insolvency.  ...  The proposed model is beneficial to policy-makers to create guidelines for the solvency regulations and roles of the government in protecting the public against insolvency.  ...  It showed very good performance for strong, medium and weak companies, 96.3% of the strong, 75.0% of the medium and 94.4% of the weak are classified correctly.  ... 
arXiv:1203.6424v2 fatcat:gob5pncdfvcnxi6wrnjt2e5wue

A New Competitive Neural Architecture for Object Classification

Mohammed Madiafi, Jamal Ezzahar, Kamal Baraka, Abdelaziz Bouroumi
2022 Applied Sciences  
The second step aims to assess the quality of the learned prototypes in terms of classification results.  ...  The first step consists in finding a set of one or more optimal prototypes for each of the c classes that form the training data.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/app12094724 fatcat:3u4io6i2qnevdlb45sc3o5vtq4

Deep Learning for Audio Transcription on Low-Resource Datasets [article]

Veronica Morfi, Dan Stowell
2018 arXiv   pre-print
Secondly, deep neural networks need a very large amount of labelled training data to achieve good quality performance, yet in practice it is difficult to collect enough samples for most classes of interest  ...  In training a deep learning system to perform audio transcription, two practical problems may arise.  ...  For a multi-class dataset, one would have to train a separate network for each class in order to perform single class event detection.  ... 
arXiv:1807.03697v2 fatcat:oivks2fplvfwzgzdentyf45osy

A comparative study of calibration methods for imbalanced class incremental learning

Umang Aggarwal, Adrian Popescu, Eden Belouadah, Celine Hudelot
2021 Multimedia tools and applications  
Existing research approaches these two problems separately while they co-occur in real world applications. Here, we study the problem of learning incrementally from imbalanced datasets.  ...  Since memory is bounded, old classes are learned with fewer images than new classes and an imbalance due to incremental learning is added to the initial dataset imbalance.  ...  Finally, inter-class separation is favored in order to better separate embeddings of past and new classes. 1 : Means and standard deviations of image counts in the original datasets (orig) and the two  ... 
doi:10.1007/s11042-020-10485-5 fatcat:ojnzzzjm45eqvl4svkwlsmbrxi
« Previous Showing results 1 — 15 out of 771,855 results