Filters








1,426 Hits in 6.9 sec

Greedy Learning of Multiple Objects in Images Using Robust Statistics and Factorial Learning

Christopher K.I. Williams, Michalis K. Titsias
2004 Neural Computation  
We consider data that are images containing views of multiple objects. Our task is to learn about each of the objects present in the images.  ...  We develop a method to extract object models sequentially from the data by making use of a robust statistical method, thus avoiding the combinatorial explosion, and present results showing successful extraction  ...  Acknowledgments C.W. thanks Geoff Hinton for helpful discussions concerning the idea of learning one object at a time, Zoubin Ghahramani for helpful discussions on the GREEDY algorithm, and Andrew Fitzgibbon  ... 
doi:10.1162/089976604773135096 pmid:15070509 fatcat:j6cy2sxkevdk5hvp3uc3tjah44

Learning About Multiple Objects in Images: Factorial Learning without Factorial Search

Christopher K. I. Williams, Michalis K. Titsias
2002 Neural Information Processing Systems  
We consider data which are images containing views of multiple objects. Our task is to learn about each of the objects present in the images.  ...  We develop a method to extract object models sequentially from the data by making use of a robust statistical method, thus avoiding the combinatorial explosion, and present results showing successful extraction  ...  Acknowledgements: CW thanks Geoff Hinton for helpful discussions concerning the idea of learning one object at a time.  ... 
dblp:conf/nips/WilliamsT02 fatcat:2t6wv4c5dbhcjiixsd6zxy6kem

Deep Belief Nets [chapter]

Geoffrey I. Webb, Johannes Fürnkranz, Johannes Fürnkranz, Johannes Fürnkranz, Geoffrey Hinton, Claude Sammut, Joerg Sander, Michail Vlachos, Yee Whye Teh, Ying Yang, Dunja Mladeni, Janez Brank (+6 others)
2011 Encyclopedia of Machine Learning  
-The object is centered.-The edges of the image are mainly blank.-The background is uniform and bright. • To make learning faster I used simplified the data:-Throw away one image.  ...  Why does greedy learning fail in a directed module?  ...  • Images typically have strong pair-wise correlations. • Learning higher order statistics is difficult when there are strong pair-wise correlations.  ... 
doi:10.1007/978-0-387-30164-8_208 fatcat:2sh6wfs5c5cwhitpraqhez7oxa

Deep Reinforcement Learning for Distributed Flow Shop Scheduling with Flexible Maintenance

Qi Yan, Wenbin Wu, Hongfeng Wang
2022 Machines  
With the introduction of machine maintenance constraints in multi-factory production scheduling, the complexity and computation time of solving the problem increases substantially in large-scale arithmetic  ...  The proposed solution framework is compared to the DQN with fixed greedy rate, in addition to two well-known metaheuristic algorithms, including the genetic algorithm and the iterated greedy algorithm.  ...  [29] developed robust and resilient scheduling approaches in a multi-factory network with periodic maintenance and uncertain disturbances, in which the proposed model in small and medium instances were  ... 
doi:10.3390/machines10030210 fatcat:qsg6cfy5dfhkljczifke3brxly

Selection and context for action recognition

Dong Han, Liefeng Bo, Cristian Sminchisescu
2009 2009 IEEE 12th International Conference on Computer Vision  
, but also because of subtle behavioral patterns among interacting people or between people and objects in images.  ...  In this paper we present contextual scene descriptors and Bayesian multiple kernel learning methods for recognizing human action in complex non-instrumented video.  ...  Image and Video Descriptors In this section we describe the features used in experiments.  ... 
doi:10.1109/iccv.2009.5459427 dblp:conf/iccv/HanBS09 fatcat:k2375ejm4ngfdihcwv5lzk6jsu

Greedy Learning of Deep Boltzmann Machine (GDBM)'s Variance and Search Algorithm for Efficient Image Retrieval

Mudhafar Jalil Jassim Ghrabat, Guangzhi Ma, Hong Liu, Zaid Ameen Abduljabbar, Mustafa A.Al Sibahee, Safa Jalil Jassim
2019 IEEE Access  
Finally, the relevant features are utilized for the greedy learning of deep Boltzmann machine classifier (GDBM).  ...  Initially, a preprocessing technique is introduced in this study, a technique that uses a median filter to remove noise to achieve improved accuracy and reliability.  ...  in an image or the selected region of an image; the shape feature provides us information about structure and statistical features in an image or the selected region of an image.  ... 
doi:10.1109/access.2019.2948266 fatcat:lzhyuujngvhehm54jlue2kfis4

LETHA: Learning from High Quality Inputs for 3D Pose Estimation in Low Quality Images

Adrian Penate-Sanchez, Francesc Moreno-Noguer, Juan Andrade-Cetto, Francois Fleuret
2014 2014 2nd International Conference on 3D Vision  
We first automatically build a 3D model of the object of interest from high-definition images, and devise from it a pose-indexed feature extraction scheme.  ...  Our results demonstrate that the method combines the strengths of global image representations, discriminative even for very tiny images, and the robustness to occlusions of approaches based on local feature  ...  THE LETHA APPROACH In this section we describe an implementation of the LETHA learning paradigm, applied to the estimation of the pose of an object in low-quality images.  ... 
doi:10.1109/3dv.2014.18 dblp:conf/3dim/SanchezMAF14 fatcat:2ropuitdqbfm7juwybdib5fcl4

Online Semi-Supervised Learning with Deep Hybrid Boltzmann Machines and Denoising Autoencoders [article]

Alexander G. Ororbia II, C. Lee Giles, David Reitter
2016 arXiv   pre-print
Theoretical motivations and algorithms for joint learning for each are presented. We apply the new models to the domain of data-streams in work towards life-long learning.  ...  Two novel deep hybrid architectures, the Deep Hybrid Boltzmann Machine and the Deep Hybrid Denoising Auto-encoder, are proposed for handling semi-supervised learning problems.  ...  ACKNOWLEDGMENTS We would like to thank Hugo Larochelle for useful conversations that helped inform this work. The shortcomings of this paper, however, are ours and ours alone.  ... 
arXiv:1511.06964v7 fatcat:iupnmfuxjnflbekwfh7ulfnzfe

Effect of human guidance and state space size on Interactive Reinforcement Learning

Halit Bener Suay, Sonia Chernova
2011 2011 RO-MAN  
We present the first study of Interactive Reinforcement Learning in realworld robotic systems.  ...  The Interactive Reinforcement Learning algorithm enables a human user to train a robot by providing rewards in response to past actions and anticipatory guidance to guide the selection of future actions  ...  This mapping is determined based on the number of Speeded Up Robust Features (SURF) [17] identified in the image of the object and identifies the object as either a solid colored ball or a character magnet  ... 
doi:10.1109/roman.2011.6005223 dblp:conf/ro-man/SuayC11 fatcat:numbqfpuyjd4hoxshtckfasdtq

Adaptive Object Tracking via Multi-Angle Analysis Collaboration

Wanli Xue, Zhiyong Feng, Chao Xu, Zhaopeng Meng, Chengwei Zhang
2018 Sensors  
Concretely, the tracker, regarded as an agent, is trained with Q-learning algorithm and ϵ -greedy exploration strategy, where we adopt a customized rewarding function to encourage robust object tracking  ...  Numerous contrast experimental evaluations on the OTB50 benchmark demonstrate the effectiveness of the strategies and improvement in speed and accuracy of MACT tracker.  ...  The learning of strategy is completed by the Q-learning and -greedy exploration in reinforcement learning.  ... 
doi:10.3390/s18113606 fatcat:c3xeiuhk3bfwlgabmsxjf6t6ai

Dictionary Learning for Stereo Image Representation

Ivana Tošić, Pascal Frossard
2011 IEEE Transactions on Image Processing  
The ML objective function is optimized using the expectation-maximization algorithm.  ...  We apply the learning algorithm to the case of omnidirectional images, where we learn scales of atoms in a parametric dictionary.  ...  ACKNOWLEDGMENT The authors would like to thank the members of the Redwood Center for Theoretical Neuroscience at UC Berkeley, for the fruitful discussions on ML dictionary learning.  ... 
doi:10.1109/tip.2010.2081679 pmid:20889431 fatcat:5pupffvotbefza2lwybyw2usqi

Deep learning for sensor-based activity recognition: A Survey

Jindong Wang, Yiqiang Chen, Shuji Hao, Xiaohui Peng, Lisha Hu
2018 Pattern Recognition Letters  
Recently, the recent advancement of deep learning makes it possible to perform automatic high-level feature extraction thus achieves promising performance in many areas.  ...  This paper surveys the recent advance of deep learning based sensor-based activity recognition. We summarize existing literature from three aspects: sensor modality, deep model, and application.  ...  Moreover, most of the existing PR models mainly focus on learning from static data; whereas activity data in real life are coming in stream, which requires robust online or incremental learning schema.  ... 
doi:10.1016/j.patrec.2018.02.010 fatcat:xwtshq6ivnggjicqhn6ejzhksi

On deep generative models with applications to recognition

Marc'Aurelio Ranzato, Joshua Susskind, Volodymyr Mnih, Geoffrey Hinton
2011 CVPR 2011  
The most popular way to use probabilistic models in vision is first to extract some descriptors of small image patches or object parts using well-engineered features, and then to use statistical learning  ...  In this work, we use one of the best, pixel-level, generative models of natural images -a gated MRF -as the lowest level of a deep belief network (DBN) that has several hidden layers.  ...  They also ackowledge the use of the CUDAMat library [16] to expedite training by using GPUs. The research was funded by grants from NSERC, CFI and CIFAR and by gifts from Google and Microsoft.  ... 
doi:10.1109/cvpr.2011.5995710 dblp:conf/cvpr/RanzatoSMH11 fatcat:n5ef527oy5e6lb5fmfyqhp6ydi

Deep Learning for Environmentally Robust Speech Recognition: An Overview of Recent Developments [article]

Zixing Zhang, Jürgen Geiger, Jouni Pohjalainen, Amr El-Desoky Mousa, Wenyu Jin, Björn Schuller
2018 arXiv   pre-print
those involved in the development of environmentally robust speech recognition systems.  ...  In this light, we review recently developed, representative deep learning approaches for tackling non-stationary additive and convolutional degradation of speech with the aim of providing guidelines for  ...  Standard Corpora and Evaluation Metrics To better compare the effectiveness of various deep learning approaches for noise-robust ASR, we introduce a set of widely used standard databases (see TableI) in  ... 
arXiv:1705.10874v3 fatcat:evdhqnj7eraa5jiolakuf4mf3e

A Review of Artificial Intelligence Algorithms Used for Smart Machine Tools

Chih-Wen Chang, Hau-Wei Lee, Chein-Hung Liu
2018 Inventions  
learning (SL) scheme [62], a weakly SL algorithm and high-level feature learning for object detection in remotely sensed optical images [63], the introduction of an off-policy reinforcement learning (  ...  for the extraction and composition of robust features [53], a large-scale deep unsupervised learning (UL) scheme for graphics processors and the building of high-level features [54,55], unsupervised learning  ...  weakly labeled images and start iterative learning from the object detector.  ... 
doi:10.3390/inventions3030041 fatcat:6qrwhmrl2bfwrgmovqvsyx5p3y
« Previous Showing results 1 — 15 out of 1,426 results