Abstract | ||
---|---|---|
This paper introduces RankMap, a platform-aware end-to-end framework for efficient execution of a broad class of iterative learning algorithms for massive and dense data sets. Our framework exploits data structure to scalably factorize it into an ensemble of lower rank subspaces. The factorization creates sparse low-dimensional representations of the data, a property which is leveraged to devise e... |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/TNNLS.2016.2631581 | IEEE Transactions on Neural Networks and Learning Systems |
Keywords | Field | DocType |
Matrix decomposition,Distributed databases,Computational modeling,Iterative algorithms,Sparse matrices,Signal processing algorithms,Partitioning algorithms | Data structure,Data set,Scheduling (computing),Computer science,Server,Theoretical computer science,Low-rank approximation,Bandwidth (signal processing),Artificial intelligence,Iterative learning control,Power iteration,Machine learning | Journal |
Volume | Issue | ISSN |
29 | 7 | 2162-237X |
Citations | PageRank | References |
1 | 0.35 | 0 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Azalia Mirhoseini | 1 | 238 | 18.68 |
Eva L. Dyer | 2 | 73 | 6.97 |
Ebrahim M. Songhori | 3 | 106 | 8.05 |
Richard G. Baraniuk | 4 | 5053 | 489.23 |
Farinaz Koushanfar | 5 | 3055 | 268.84 |