Filters








7 Hits in 4.8 sec

Stochastic Recursive Gradient Support Pursuit and Its Sparse Representation Applications

Fanhua Shang, Bingkun Wei, Yuanyuan Liu, Hongying Liu, Shuang Wang, Licheng Jiao
2020 Sensors  
In addition, some stochastic hard thresholding methods were also proposed, such as stochastic gradient hard thresholding (SG-HT) and stochastic variance reduced gradient hard thresholding (SVRGHT).  ...  In recent years, a series of matching pursuit and hard thresholding algorithms have been proposed to solve the sparse representation problem with ℓ0-norm constraint.  ...  Acknowledgments: We thank all the reviewers for their valuable comments. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/s20174902 pmid:32872609 pmcid:PMC7506682 fatcat:wbpkarfakrapxoyphzkp6lao7u

On Seven Fundamental Optimization Challenges in Machine Learning [article]

Konstantin Mishchenko
2021 arXiv   pre-print
The exchange of ideas between these fields has worked both ways, with machine learning building on standard optimization procedures such as gradient descent, as well as with new directions in the optimization  ...  The fifth challenge is the development of an algorithm for distributed optimization with quantized updates that preserves linear convergence of gradient descent.  ...  Above all, with Peter Richtárik's help, I quickly became able to work independently and collaborate with people from different countries and backgrounds, which lays a solid foundation for my future work  ... 
arXiv:2110.12281v1 fatcat:c4oc7xv6fvdqdik4hwegrcnsqm

The Internet of Federated Things (IoFT): A Vision for the Future and In-depth Survey of Data-driven Approaches for Federated Learning [article]

Raed Kontar, Naichen Shi, Xubo Yue, Seokhyun Chung, Eunshin Byon, Mosharaf Chowdhury, Judy Jin, Wissam Kontar, Neda Masoud, Maher Noueihed, Chinedum E. Okwudire, Garvesh Raskutti (+3 others)
2021 arXiv   pre-print
This article provides a vision for IoFT and a systematic overview of current efforts towards realizing this vision.  ...  model that quickly adapts to new devices or learning tasks.  ...  Federated ing latin hypercube design for sequential sampling of accelerated stochastic gradient descent. In 34th computer experiments.  ... 
arXiv:2111.05326v1 fatcat:bbgdhtuqcrhstgakt2vxuve2ca

On Seven Fundamental Optimization Challenges in Machine Learning

Konstantin Mishchenko
2021
The exchange of ideas between these fields has worked both ways, with ' learning building on standard optimization procedures such as gradient descent, as well as with new directions in the optimization  ...  The fifth challenge we resolve in the affirmative is the development of an algorithm for distributed optimization with quantized updates that preserves global linear convergence of gradient descent.  ...  ACKNOWLEDGEMENTS The last inequality is trivially satisfied for all k ≥ 0. We are not ready to prove Theorem 6.5.6. Below we repeat its statement and prove the proof right afterwards.  ... 
doi:10.25781/kaust-nv6ui fatcat:2ooj4nwi5bbcjbswy4ebl5hkhm

Computational Methods in Drug Discovery

G. Sliwoski, S. Kothiwale, J. Meiler, E. W. Lowe
2013 Pharmacological Reviews  
HipHop is the simpler of the two algorithms and looks for common 3D arrangements of features only for compounds with a threshold activity against the target.  ...  This is less error prone to overfitting and requires smaller datasets to begin with.  ... 
doi:10.1124/pr.112.007336 pmid:24381236 pmcid:PMC3880464 fatcat:4dzrdkspkjecnombnchznma2ny

Data-Driven Approaches to Exploratory Visual Analysis of Mass Spectrometry Imaging Data

Karsten Wüllems
2021
By using the spatial positions associated with each spectrum, the similarity values can be visualized on the sample by pseudo-coloring.  ...  that similarity relationships can be found that cannot be detected when analyzing single molecular images (m{z-images) [Wüllems et al., in revision]. xiii 6 Combining the Spatial and Spectral Domain for  ...  The low-dimensional graph representation is optimized by using cross entropy as an objective function and stochastic gradient descent as an optimizer [75, 123] .  ... 
doi:10.4119/unibi/2956953 fatcat:7ljegttohzfc5gylqkybnxi5w4

New algorithms for Quadratic Unconstrained Binary Optimization (QUBO) with applications in engineering and social sciences

Gabriel Tavares
2008
First, there is an algorithmic aspect that tells how to preprocess the problem, how to find heuristics, how to get improved bounds and how to solve the problem with all the above ingredients.Second, there  ...  QUBO are analyzed, which include a novel probabilistic based class of methods.It is shown that there is a unique maximal set of persistencies for the linearization model for QUBO.This set is determined  ...  Γ 1 is used for a term with less one variable and coefficient +1.  ... 
doi:10.7282/t3xk8fs2 fatcat:q4wvuff66bgidcykqks3o477pq