Filters








30,826 Hits in 7.9 sec

Super-efficiency of automatic differentiation for functions defined as a minimum [article]

Pierre Ablin, Gabriel Peyré, Thomas Moreau
2020 arXiv   pre-print
In min-min optimization or max-min optimization, one has to compute the gradient of a function defined as a minimum.  ...  We find that the error of the automatic estimator is close to the square of the error of the analytic estimator, reflecting a super-efficiency phenomenon.  ...  This work was funded in part by the French government under management of Agence Nationale de la Recherche as part of the "Investissements d'avenir" program, reference ANR-19-P3IA-0001 (PRAIRIE 3IA Institute  ... 
arXiv:2002.03722v1 fatcat:bumyy323vfexxamteyq3nmez2a

Automatic Zig-Zag sampling in practice [article]

Alice Corbella, Simon E F Spencer, Gareth O Roberts
2022 arXiv   pre-print
., correct invariant distribution, ergodicity, and super-efficiency).  ...  Lastly, we demonstrate that the super-efficiency property, i.e. the ability to draw one independent sample at a lesser cost than evaluating the likelihood of all the data, can be obtained in practice.  ...  The authors are grateful to professor Jonathan Tawn for his insights and comments on the extreme-values methods used in Section 6. Disclosures and Declarations  ... 
arXiv:2206.11410v1 fatcat:fyk643hgercfzcqla7gsnam2uu

Automatic Differentiation in Intelligent Tutoring Systems [chapter]

Vladislav Katkov, Ada Novosselova
2002 Networking the Learner  
The expansion of languages C/C++, Fortran-77 and Pascal for these purposes and its realisarlon as system Diff are described.  ...  The automarle differenrlarlon is widely used with the decision of a wide class of problems.  ...  The program CODEX and its further modification SUPER-CODEX intended for automatic differentiation on a Fortran were developed at the University of Wisconsin-Madison (USA) under a management of Reiter  ... 
doi:10.1007/978-0-387-35596-2_45 fatcat:wxpeiyebxjfofgyufnnfgluqf4

Generalized reaction–diffusion equations

G.W. Wei
1999 Chemical Physics Letters  
A number of super-diffusion operators are introduced for fast and effective smoothing. Statistical information is utilized for robust edge-stopping and diffusion-rate estimation.  ...  A quasi-interpolating wavelet algorithm is utilized for numerical computations. Computer experiments indicate that the present algorithm is efficient for edge detecting and noise removing.  ...  Acknowledgements This work was supported in part by the National University of Singapore. The author thanks Drs. E.C. Chang and Y.P. Wang, and Prof. Yingfei Yi for useful discussions.  ... 
doi:10.1016/s0009-2614(99)00270-5 fatcat:lf4tphlvafba7fdd533cnwzmbe

FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search [article]

Bichen Wu, Xiaoliang Dai, Peizhao Zhang, Yanghan Wang, Fei Sun, Yiming Wu, Yuandong Tian, Peter Vajda, Yangqing Jia, Kurt Keutzer
2019 arXiv   pre-print
FBNets, a family of models discovered by DNAS surpass state-of-the-art models both designed manually and generated automatically.  ...  Designing accurate and efficient ConvNets for mobile devices is challenging because the design space is combinatorially large.  ...  Instead of solving for the optimal architecture aA, which has a discrete search space, we relax the problem to optimize the probability P θ of the stochastic super net to achieve the minimum expected  ... 
arXiv:1812.03443v3 fatcat:eelcfkt3hzfyzjj7cfo6ydn2sa

Automatic Search Dense Connection Module for Super-Resolution

Huaijuan Zang, Guoan Cheng, Zhipeng Duan, Ying Zhao, Shu Zhan
2022 Entropy  
For this purpose, we present a lightweight network that automatically searches dense connection (ASDCN) for image super-resolution (SR), which effectively reduces redundancy in dense connection and focuses  ...  The development of display technology has continuously increased the requirements for image resolution.  ...  Conclusions In this paper, we propose a framework for automatically searching dense connection modules for single image super-resolution.  ... 
doi:10.3390/e24040489 pmid:35455153 pmcid:PMC9030154 fatcat:gxugmaxlmrcrjjupiyvo32tpmi

FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search

Bichen Wu, Kurt Keutzer, Xiaoliang Dai, Peizhao Zhang, Yanghan Wang, Fei Sun, Yiming Wu, Yuandong Tian, Peter Vajda, Yangqing Jia
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Differentiable neural architecture search (DNAS) for ConvNet design. DNAS explores a layer-wise space that each layer of a ConvNet can choose a different block.  ...  FBNets (Facebook-Berkeley-Nets), a family of models discovered by DNAS surpass state-of-the-art models both designed manually and generated automatically.  ...  Instead of solving for the optimal architecture aA , which has a discrete search space, we relax the problem to optimize the probability P θ of the stochastic super net to achieve the minimum expected  ... 
doi:10.1109/cvpr.2019.01099 dblp:conf/cvpr/WuDZWSWTVJK19 fatcat:eyxymk3kwnc3jnonjunf7nmahy

AbstractDifferentiation.jl: Backend-Agnostic Differentiable Programming in Julia [article]

Frank Schäfer, Mohamed Tarek, Lyndon White, Chris Rackauckas
2022 arXiv   pre-print
As a response to these issues, we present AbstractDifferentiation.jl for the automatized generation of an extensive, unified, user-facing API for any AD package.  ...  No single Automatic Differentiation (AD) system is the optimal choice for all problems.  ...  ∂P is (commonly) realized by automatic differentiation (AD), a family of techniques to efficiently and accurately differentiate numeric functions expressed as computer programs.  ... 
arXiv:2109.12449v2 fatcat:vp3u4734vnfdbevit27czlvuhq

Using Low-Density Parity-Check Codes to Reduce the Effect of Laser Line width For Optical Communication

R. S. Fyath, Hussam A. Darweesh
2011 مجلة النهرين للعلوم الهندسية  
This paper addresses the application of the Low-Density Parity-Check (LDPC) codes as Forward Error Correcting (FEC) codes to relax the laser linewidth requirement.  ...  The performance of coherent optical communication systems is degraded significantly by the phase noise of the semiconductor lasers.  ...  of a local code C 0 (n, k) as shown [14] In general, one seeks LDPC codes for which the local codes C 0 (n, k) has large minimum distance (d min ) and a code rate as high as possible.  ... 
doaj:0f34c7e8ebfc4653bd27c9b0ba99bd2c fatcat:nyt5qsu7gbbdbb5tz7g6q6jfnu

Automatic coarse graining of polymers

Roland Faller
2004 Polymer  
Several ways to obtain a time-scale for dynamic mapping are discussed additionally. Finally, similarities to other simulation areas where automatic optimization are applied as well are pointed out.  ...  All these techniques derive effective potentials for multi--atom units or super--atoms from atomistic simulations.  ...  I especially want to thank Qi Sun for some analysis of the cis-PI system.  ... 
doi:10.1016/j.polymer.2003.11.053 fatcat:hbaa32fa7rdtncnatungelskzu

Gradient descent with momentum — to accelerate or to super-accelerate? [article]

Goran Nakerst, John Brennan, Masudul Haque
2020 arXiv   pre-print
Considering a one-parameter quadratic loss function, the optimal value of the super-acceleration can be exactly calculated and analytically estimated.  ...  We consider gradient descent with 'momentum', a widely used method for loss function minimization in machine learning.  ...  (a) The time scale T for reaching the minimum of L(θ) is a non-monotonic function of the super-acceleration parameter σ, and has a well-defined minimum.  ... 
arXiv:2001.06472v1 fatcat:bmrninyrxna47a4fjo763n5ojm

Efficient discovery of risk patterns in medical data

Jiuyong Li, Ada Wai-chee Fu, Paul Fahey
2009 Artificial Intelligence in Medicine  
The proposed method does not discover a large number of uninteresting superfluous patterns as an association mining approach does.  ...  Objective: This paper studies a problem of efficiently discovering risk patterns in medical data.  ...  We thank Dr Mark Laverty for providing the opportunity for the case study.  ... 
doi:10.1016/j.artmed.2008.07.008 pmid:18783927 fatcat:nrfzlexmpnaupbtthcqszhdy3u

Optimality conditions for various efficient solutions involving coderivatives: From set-valued optimization problems to set-valued equilibrium problems

Truong Xuan Duc Ha
2012 Nonlinear Analysis  
Secondly, we show that there is an equivalence between solutions of SOP and solutions of SEP and based on this equivalence we derive similar results for solutions of SEP as well as for solutions of a vector  ...  Firstly, we obtain in a unified scheme scalar characterization with the help of Hiriart-Urruty's signed distance function, necessary/sufficient conditions for weakly, strongly, properly efficient solutions  ...  This work was supported in part by a grant of the National Foundation for Science and Technology Development, Vietnam.  ... 
doi:10.1016/j.na.2011.03.015 fatcat:fh5ksewoh5hbbj6ssccboob3se

Segmentation of Defects in Textile Images Based on Adaptive Sparse Domain Selection Method

M Fathu Nisha, L Ganesh Babu, P Vasuki, S Mohamed Mansoor Roomi
2019 Zenodo  
Generalized fabric inspection system needs perfect and efficient pattern recognition to detect and identify different types of defects in various kinds of fabrics.  ...  In real time production environment, it is difficult to detect defects automatically for whole fabrics, although fabrics may be characterized by color, pattern, texture, and material quality.  ...  Precision defined as the percentage of correct alarm during detection, sensitivity defined as the percentage of defective samples that are correctly identified, specificity defined as the percentage of  ... 
doi:10.5281/zenodo.3573569 fatcat:eykcf7gwbjflpjytjwqitwwt4q

Fully automatic segmentation and objective assessment of atrial scars for long-standing persistent atrial fibrillation patients using late gadolinium-enhanced MRI

Guang Yang, Xiahai Zhuang, Habib Khan, Shouvik Haldar, Eva Nyktari, Lei Li, Ricardo Wage, Xujiong Ye, Greg Slabaugh, Raad Mohiaddin, Tom Wong, Jennifer Keegan (+1 others)
2018 Medical Physics (Lancaster)  
In this study, we proposed a novel fully automatic pipeline to achieve an accurate and objective atrial scarring segmentation and assessment of LGE MRI scans for the AF patients.  ...  to LGE MRI, and (2) a super-pixel and supervised learning based approach to delineate the distribution and extent of atrial scarring in LGE MRI.  ...  FEP is defined as the volume of scar tissue as a percentage of the atrial wall volume. 15 2.E.5.  ... 
doi:10.1002/mp.12832 pmid:29480931 pmcid:PMC5969251 fatcat:adpi4ouwbveh5dghpkeidoqory
« Previous Showing results 1 — 15 out of 30,826 results