Filters








13,387 Hits in 4.0 sec

Open Set Domain Adaptation by Backpropagation [article]

Kuniaki Saito, Shohei Yamamoto, Yoshitaka Ushiku, Tatsuya Harada
2018 arXiv   pre-print
In this paper, we propose a method for an open set domain adaptation scenario which utilizes adversarial training.  ...  A target domain can contain samples of classes that are not shared by the source domain.  ...  Left: Existing setting of open set domain adaptation Sour ce Backpack Bicycle Mug Source Target Backpack Bicycle Mug Unknown Unknown Open Set Domain Adaptation proposed by Busto et al.  ... 
arXiv:1804.10427v2 fatcat:wwjncsubovgornp2khfrsg74s4

Open Set Domain Adaptation by Backpropagation [chapter]

Kuniaki Saito, Shohei Yamamoto, Yoshitaka Ushiku, Tatsuya Harada
2018 Lecture Notes in Computer Science  
In this paper, we propose a method for an open set domain adaptation scenario, which utilizes adversarial training.  ...  However, in practice, a target domain can contain samples of classes that are not shared by the source domain.  ...  Left: Existing setting of open set domain adaptation Fig. 2 . 2 (a): Closed set domain adaptation with distribution matching method.  ... 
doi:10.1007/978-3-030-01228-1_10 fatcat:rnrhixy54zb7ljrwfcu3fyskya

An Extensible and Modular Design and Implementation of Monte Carlo Tree Search for the JVM [article]

Larkin Liu, Jun Tao Luo
2021 arXiv   pre-print
We demonstrate that the design of the MCTS implementation provides ease of adaptation for unique heuristics and customization across varying Markov Decision Process (MDP) domains.  ...  We define key class abstractions allowing the MCTS library to flexibly adapt to any well defined Markov Decision Process or turn-based adversarial game.  ...  This has been shown to work well for adversarial games to avoid traps set by the opposite player.  ... 
arXiv:2108.10061v1 fatcat:phtyonwkrfhxjg4z524bmb6k3q

Intelligent Fault Diagnosis for Bearing Dataset Using Adversarial Transfer Learning based on Stacked Auto-Encoder

Jipu Li, Ruyi Huang, Weihua Li
2020 Procedia Manufacturing  
Therefore, inspired by the idea of open set domain adaptation, an adversarial transfer learning based on stacked auto-encoder method is proposed to address new fault emerging problem for the target domain  ...  Therefore, inspired by the idea of open set domain adaptation, an adversarial transfer learning based on stacked auto-encoder method is proposed to address new fault emerging problem for the target domain  ...  Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant No.51875208) and National Key R&D Program of China (Grant No.2018YFB1702402).  ... 
doi:10.1016/j.promfg.2020.06.014 fatcat:6tkpxeecyjeqlggidpptxyxar4

A Recap of Early Work on Theory and Knowledge Refinement

Raymond J. Mooney, Jude W. Shavlik
2021 AAAI Spring Symposia  
Not all the facts about the domain at hand may be referenced by the rule set (these are the open red circles on the bottom), but an important role for them might be discovered during training. 2.  ...  Similarly, some rule antecedents might be pushed toward zero by backpropagation, essentially removing them (backpropagation also converts the Boolean algebra of rule sets into weighted sums that are input  ... 
dblp:conf/aaaiss/MooneyS21 fatcat:kwjw4wfjd5bblmtbdvdo7367i4

Application ofArtificial Neural Networking for Determining the Plane of Vibration in Rotating System

Arka Sen, Manik Chandra Majumder, Sumit Mukhopadhyay, Robin Kumar Biswas, Subhadeep Roy
2017 IOSR Journal of Mechanical and Civil Engineering  
, where by directly feeding the RMS and Phase values of vibration, the unbalance plane can be detected with minimum error.  ...  In this paper a new approach for Artificial Neural Networking using Feed Forward Back Propagation Method and Levenberg-Marquardt backpropagation training function has been developed using Java Programming  ...  (Gradient descent with momentum backpropagation) Neural network by TRAINGDA (Gradient descent with adaptive lr backpropagation) Neural network by TRAINGDX (Gradient descent w/momentum & adaptive lr backpropagation  ... 
doi:10.9790/1684-1401052335 fatcat:r5vzjvne7zbynooh3lqhhphhje

mctreesearch4j: A Monte Carlo Tree Search Implementation for the JVM

Larkin Liu, Jun Luo
2022 Journal of Open Source Software  
In addition, key class abstractions are designed for the library to flexibly adapt to any well-defined Markov Decision Process (MDP) or turn-based adversarial games.  ...  Furthermore, mctreesearch4j is capable of customization across a variety of MDP domains, consequently enabling the adoption of MCTS heuristics and customization into the core library with ease.  ...  . • Adaptability: Adaptability is defined as the ability for MDP domain to be easily integrated into the mctreesearch4j framework using provided class abstractions.  ... 
doi:10.21105/joss.03804 fatcat:4py4kcnxrrhwzbej3nswupvyea

Accented Speech Recognition Based on End-to-End Domain Adversarial Training of Neural Networks

Hyeong-Ju Na, Jeong-Sik Park
2021 Applied Sciences  
The DANN plays a role as a domain adaptation in which the training data and test data have different distributions.  ...  Thus, our approach is expected to construct a reliable ASR model for accented speech by reducing the distribution differences between accented speech and standard speech.  ...  When the gradient of the domain classifier ϑL d ϑθ d is backpropagated into the downstream of the ASR model, the gradient is multiplied by λ, which can be set between 0 and 1.  ... 
doi:10.3390/app11188412 fatcat:qawsqisvqjfsdh4rc4iifp7ejq

Towards MCTS for Creative Domains

Cameron Browne
2011 International Conference on Computational Creativity  
The primary application of this extended MCTS model will be for creative domains, as it maps naturally to a range of procedural content generation tasks for which Markovian or evolutionary approaches would  ...  Acknowledgements This work is supported by EPSRC standard grant EP/ I001964. Thanks to Simon Colton, Stephen Tavener, Phillip Rohlfshagen, Greg Schmidt and the anonymous reviewers for useful comments.  ...  MCTS is also: Aheuristic: No heuristic domain knowledge is required. Asymmetric: The search adapts to fit the search space. Convergent: The search converges to optimal solutions.  ... 
dblp:conf/icccrea/Browne11 fatcat:sevenwcsuzfodlj6dq6sxbxt4u

Solving Multiclass Learning Problems via Error-Correcting Output Codes

T. G. Dietterich, G. Bakiri
1995 The Journal of Artificial Intelligence Research  
Thedefinition is acquired by studying collections of training examples ofthe form [x_i, f (x_i)].  ...  We show that these outputrepresentations improve the generalization performance of both C4.5and backpropagation on a wide range of multiclass learning tasks.  ...  In most domains, we used the extremely fast backpropagation implementation provided by the CNAPS neurocomputer Adaptive Solutions, 1992.  ... 
doi:10.1613/jair.105 fatcat:tyo6yvwqdzcz7oar2nxp3etchq

Solving Multiclass Learning Problems via Error-Correcting Output Codes [article]

T. G. Dietterich, G. Bakiri
1995 arXiv   pre-print
The definition is acquired by studying collections of training examples of the form [x_i, f (x_i)].  ...  Multiclass learning problems involve finding a definition for an unknown function f(x) whose range is a discrete set containing k > 2 values (i.e., k "classes").  ...  In most domains, we used the extremely fast backpropagation implementation provided by the CNAPS neurocomputer (Adaptive Solutions, 1992) .  ... 
arXiv:cs/9501101v1 fatcat:fvi6ddbhlnd4rhmmw6yewcteji

A Diagnosis Framework for High-reliability Equipment with Small Sample Based on Transfer Learning

Jinxin Pan, Bo Jing, Xiaoxuan Jiao, Shenglong Wang, Qingyi Zhang, Kiyong Oh
2022 Complexity  
For adequate fitness, the joint adaptation of conditional distribution and marginal distribution was used between the two domains.  ...  As a result, the fault diagnosis rate increases by 28.6% through our proposed model, which is more precise than other classical methods.  ...  In other words, the network trained by 2 sets of data was verified by 3494 sets of validation data to judge the diagnostic accuracy of the network.  ... 
doi:10.1155/2022/4598725 fatcat:nwntddahbjc43fh5i26ojaur7i

Domain Adversarial Transfer Learning for Generalized Tool Wear Prediction

Peng (Edward) Wang, Matthew Russell
2020 Proceedings of the Annual Conference of the Prognostics and Health Management Society, PHM  
., regression or classification accuracy) across the labeled training examples from the source domain while maximizing the loss of the domain classifier across the source and target data sets (i.e., maximizing  ...  As a promising solution to address these challenges, Transfer Learning (TL) enables DL networks trained on a source domain and task to be applied to a separate target domain and task.  ...  . (2017) , and conditional adversarial domain adaptation discussed by Long et al. (2018) and related work by Hoffman et al. (2017) .  ... 
doi:10.36001/phmconf.2020.v12i1.1137 fatcat:j4rpmcee6ncnvng74zxbu4n2xu

Cycle Label-Consistent Networks for Unsupervised Domain Adaptation [article]

Mei Wang, Weihong Deng
2022 arXiv   pre-print
Domain adaptation aims to leverage a labeled source domain to learn a classifier for the unlabeled target domain with a different distribution.  ...  In this paper, we propose a simple yet efficient domain adaptation method, i.e.  ...  ACKNOWLEDGMENTS This work was partially supported by National Key R&D Program of China (2019YFB1406504) and BUPT Excellent Ph.D. Students Foundation CX2020207.  ... 
arXiv:2205.13957v1 fatcat:wozgv6xx6barta2hns5amjlfsi

A Cold Start Context-Aware Recommender System for Tour Planning Using Artificial Neural Network and Case Based Reasoning

Zahra Bahramian, Rahim Ali Abbaspour, Christophe Claramunt
2017 Mobile Information Systems  
This makes it difficult for the user to search for some specific information such as selecting a tour in a given city as an ordered set of points of interest.  ...  each training algorithm including gradient descent with momentum and adaptive learning rate backpropagation (GDX), resilient backpropagation (RP), conjugate gradient backpropagation with Fletcher-Reeves  ...  The algorithm has been implemented by an Android-based prototype. Data Set.  ... 
doi:10.1155/2017/9364903 fatcat:p4rvyfswbjbmzplrgbkhw2gvki
« Previous Showing results 1 — 15 out of 13,387 results