Zobrazeno 1 - 10
of 11
pro vyhledávání: '"Sajad Fathi Hafshejani"'
Modified Step Size for Enhanced Stochastic Gradient Descent: Convergence and Experiments
Publikováno v:
Mathematics Interdisciplinary Research, Vol 9, Iss 3, Pp 237-253 (2024)
This paper introduces a novel approach to enhance the performance of the stochastic gradient descent (SGD) algorithm by incorporating a modified decay step size based on $\frac{1}{\sqrt{t}}$. The proposed step size integrates a logarithmic t
Externí odkaz:
https://doaj.org/article/b7233bfe54f24bb698c14cf72b658fad
Publikováno v:
Algorithms, Vol 17, Iss 7, p 303 (2024)
A new two-step interior point method for solving linear programs is presented. The technique uses a convex combination of the auxiliary and central points to compute the search direction. To update the central point, we find the best value for step s
Externí odkaz:
https://doaj.org/article/bf0484c9218f4cb4b1ecbddcd7846736
Publikováno v:
Machine Learning with Applications, Vol 10, Iss , Pp 100440- (2022)
Low-rank matrix factorization problems such as non negative matrix factorization (NMF) can be categorized as a clustering or dimension reduction technique. The latter denotes techniques designed to find representations of some high dimensional datase
Externí odkaz:
https://doaj.org/article/943db972c2bb452bb5865e31f1de3a42
We give an improved non-monotone line search algorithm for stochastic gradient descent (SGD) for functions that satisfy interpolation conditions. We establish theoretical convergence guarantees for the algorithm for strongly convex, convex and non-co
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::628431126b58a51860b51b399876e4a5
https://doi.org/10.21203/rs.3.rs-2285238/v1
https://doi.org/10.21203/rs.3.rs-2285238/v1
Publikováno v:
Communications in Computer and Information Science ISBN: 9789819916412
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::83e245a350e5729764fe72e7f0613e6b
https://doi.org/10.1007/978-981-99-1642-9_3
https://doi.org/10.1007/978-981-99-1642-9_3
Low-rank matrix factorization problems such as non negative matrix factorization (NMF) can be categorized as a clustering or dimension reduction technique. The latter denotes techniques designed to find representations of some high dimensional datase
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::107cb0138a39aad9c401310ab826c2cb
https://doi.org/10.36227/techrxiv.12199026
https://doi.org/10.36227/techrxiv.12199026
Publikováno v:
Journal of Applied Mathematics and Computing. 62:281-300
Kernel functions play an important role in the design and complexity analysis of interior point algorithms for solving convex optimization problems. They determine both search directions and the proximity measure between the iterate and the central p
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::56879967df6af3a617aee0cc6cf784c9
https://doi.org/10.1007/s12190-019-01284-9
https://doi.org/10.1007/s12190-019-01284-9
Non-negative matrix factorization (NMF) has become a popular method for representing meaningful data by extracting a non-negative basis feature from an observed non-negative data matrix. Some of the unique features of this method in identifying hidde
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b3baaeae52e0a8c44319a33cfb199689
Low-rank matrix factorization problems such as non negative matrix factorization (NMF) can be categorized as a clustering or dimension reduction technique. The latter denotes techniques designed to find representations of some high dimensional datase
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::016b21a0bc7e2440f8afd372084efe5b
https://doi.org/10.36227/techrxiv.12199026.v1
https://doi.org/10.36227/techrxiv.12199026.v1