A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Filters
Improved Moves for Truncated Convex Models
2011
Journal of machine learning research
Preserved Fulltext
We consider the problem of obtaining an approximate maximum a posteriori estimate of a discrete random field characterized by pairwise potentials that form a truncated convex model. ...
For this problem, we propose two st-MINCUT based move making algorithms that we call Range Swap and Range Expansion. ...
form a truncated convex model). ...
dblp:journals/jmlr/KumarVT11
fatcat:tklshku7znhtjewfv7ix57isre
Improved Moves for Truncated Convex Models
2008
Neural Information Processing Systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
For this problem, we propose an improved st-MINCUT based move making algorithm. ...
We consider the problem of obtaining the approximate maximum a posteriori estimate of a discrete random field characterized by pairwise potentials that form a truncated convex model. ...
of truncated convex models). ...
dblp:conf/nips/KumarT08
fatcat:ncteglczfnf5hiye7dzwgcex4u
Multi-label Moves for MRFs with Truncated Convex Priors
[chapter]
2009
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
A multi-label move, unlike expansion and swap, gives each pixel has a choice of more than two labels to switch to. In particular, we develop several multilabel moves for truncated convex priors. ...
At each iteration, a move-making algorithm makes a proposal (move) for a pixel p to either keep its old label or switch to a new label. ...
We develop and compare several new multi-label moves for energies with a truncated convex prior, and a new multi-label move for the Potts model. ...
doi:10.1007/978-3-642-03641-5_1
fatcat:mgc7ni5gsbalxd3vqkq3nsyj3y
Multi-label Moves for MRFs with Truncated Convex Priors
2011
International Journal of Computer Vision
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
A multi-label move, unlike expansion and swap, gives each pixel has a choice of more than two labels to switch to. In particular, we develop several multilabel moves for truncated convex priors. ...
At each iteration, a move-making algorithm makes a proposal (move) for a pixel p to either keep its old label or switch to a new label. ...
We develop and compare several new multi-label moves for energies with a truncated convex prior, and a new multi-label move for the Potts model. ...
doi:10.1007/s11263-011-0491-6
fatcat:e3va2t7jezetxm67zxvqe73gd4
A local O(n2) gathering algorithm
2010
Proceedings of the 22nd ACM symposium on Parallelism in algorithms and architectures - SPAA '10
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
See Figure 3 for an illustration. r i can be moved to the point of the local convex hull closest to the point shown as p in Figure 3 . ...
Now we compute a lower bound for the expected area by which r i truncates the global convex hull in the time step t in which it is first active, depending only on its internal angle of the global convex ...
doi:10.1145/1810479.1810523
dblp:conf/spaa/DegenerKH10
fatcat:p4svwiio2ra2feqfbms47v6cja
GRMA: Generalized Range Move Algorithms for the Efficient Optimization of MRFs
2016
International Journal of Computer Vision
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
However, two problems have greatly limited the applicability of range move algorithms: 1) They are limited in the energy functions they can handle (i.e., only truncated convex functions); 2) They tend ...
In this paper, we propose two generalized range move algorithms (GR-MA) for the efficient optimization of MRFs. ...
We thank Olga Veksler for her great help to this work, and we thank Pushmeet Kohli for his valuable comments. ...
doi:10.1007/s11263-016-0944-z
fatcat:rsuz3sjtdrc25klhke3ca4j7bi
Demystifying Batch Normalization in ReLU Networks: Equivalent Convex Optimization Models and Implicit Regularization
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Experiments with CIFAR image classification highlight the effectiveness of this explicit regularization for mimicking and substantially improving the performance of standard BN networks. ...
convex objective. ...
For CIFAR-100, the chosen learning rates for the Adam, Adam-Truncated, Convex, and Convex-Truncated models were 10 −5 , 10 −4 , 5 × 10 −5 , and 10 −5 , respectively. ...
arXiv:2103.01499v3
fatcat:wsurdsbwerh6faey7hjq6cdtdm
Page 930 of Journal of Sedimentary Petrology Vol. 45, Issue 4
[page]
1975
Journal of Sedimentary Petrology
Preserved Fulltext
The Internet Archive has digitized a microfilm copy of this work. It may be possible to borrow a copy for reading.
For each category the filter is truncated. In the surf zone, the filter is truncated at the coarser end; in the swash and dune areas, the filters are truncated at the finer end. ...
In a conceptual model where sand is moved from the surf, through the swash zone, and into either beach ridges or dunes, a time sequence can be established as follows: first, surf; second, swash; third, ...
Moving object detection based on non-convex RPCA with segmentation constraint
2020
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Experimental results show that the proposed model can effectively improve the accuracy of moving object detection, and has better visual effect of foreground object detection than existed methods. ...
To solve the above problems, a new non-convex rank approximate RPCA model based on segmentation constraint is proposed. ...
For example, RPCA model can separate the background and the moving object at the same time. ...
doi:10.1109/access.2020.2977273
fatcat:jsf6bdg32nd2ncgbiheizvru5i
A CONVEX OPTIMIZATION APPROACH TO THE MODE ACCELERATION PROBLEM
2002
IFAC Proceedings Volumes
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2006; you can also visit the original URL.
The file type is application/pdf.
We will show that the problem can be cast as a convex optimization problem that can be solved via linear matrix inequalities. ? ...
The purpose of this note is to introduce an alternative procedure to the mode acceleration method when the underlying structure model includes damping. ...
This allows for the problem to be cast as a convex optimization problem. ...
doi:10.3182/20020721-6-es-1901.00463
fatcat:k246xxzaond4pmrex6ijwnbvfm
A convex optimization approach to the mode acceleration problem
2004
Automatica
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2006; you can also visit the original URL.
The file type is application/pdf.
We will show that the problem can be cast as a convex optimization problem that can be solved via linear matrix inequalities. ? ...
The purpose of this note is to introduce an alternative procedure to the mode acceleration method when the underlying structure model includes damping. ...
This allows for the problem to be cast as a convex optimization problem. ...
doi:10.1016/j.automatica.2003.12.012
fatcat:oytseywjjranrgucapou66465a
Augmented Reality as a Didactic Resource for Teaching Mathematics
2020
Applied Sciences
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Finally, we design an innovative, didactic proposal for teaching the polyhedra in the third course of compulsory Secondary Education in Spain, using the resources created with the augmented reality technology ...
Technical development takes a lot of time, resources, and effort, and for implementation in the classroom, the entire technical and curricular part must be perfectly organized. ...
The AR models will facilitate visualization (they can move, rotate the marker and the polyhedron will also move with the marker) and understand concepts like faces, vertices, and edges, that will be in ...
doi:10.3390/app10072560
fatcat:nd7zsftjdnhz5mtgz5xhsyshoq
Application and adaptation of the truncated Newton method for non-convex plasticity problems
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
The stability of the truncated Newton method is demonstrated for the, highly non-convex, Peierls-Nabarro model, solved within a Finite Element framework. ...
In successful models of such complex behaviour, its origin lies within non-convexity of the governing free energy functional. ...
Acknowledgements We would like to thank Pratheek Shanthraj of the Max Planck Institute for Iron Research for useful comments on the present work. ...
arXiv:1810.13393v1
fatcat:isnm75vuazc6rcylkydibth3pq
The forming characteristic in the single-point incremental forming of a complex shape
2017
International Journal of Nanomanufacturing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Incremental sheet forming is popularly applied because of its reduced cost, flexibility, and applicability for small batch production. ...
The sheet is clamped around its periphery with a blank holder that moves synchronously with the tool in a vertical direction, and the forming tool moves along a trajectory on the convex surface of the ...
Figure 8 shows the finite element model for the ISF process for a complex shape. ...
doi:10.1504/ijnm.2017.082409
fatcat:wtkhcxpq25ehnmzjjmch3l5wqa
Truncated Max-of-Convex Models
[article]
2016
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Truncated convex models (TCM) are a special case of pairwise random fields that have been widely used in computer vision. ...
We propose a natural generalization of TCM to high-order random fields, which we call truncated max-of-convex models (TMCM). ...
Discussion We proposed a novel family of high-order random fields called truncated max-of-convex models (TMCM) which are generalization of truncated convex models (TCM). ...
arXiv:1512.07815v2
fatcat:gk6imma7j5bqnfy25f2lxyqr6i
« Previous
Showing results 1 — 15 out of 24,647 results