GroupReduce: Block-Wise Low-Rank Approximation for Neural Language Model Shrinking.
Patrick H. ChenSi SiYang LiCiprian ChelbaCho-Jui HsiehPublished in: CoRR (2018)
Keyphrases
- language model
- block wise
- low rank approximation
- singular value decomposition
- spectral clustering
- probabilistic model
- latent semantic indexing
- n gram
- information retrieval
- document retrieval
- retrieval model
- query expansion
- test collection
- vector space model
- document representation
- image classification
- data clustering
- text retrieval
- information retrieval systems
- higher order
- relevance feedback
- feature vectors
- pairwise