Learning Succinct Models: Pipelined Compression with L1-Regularization, Hashing, Elias-Fano Indices, and Quantization

4355

Knowledge Triples

Source Entity	Relationship	Target Entity
Compression Task	uses	Feature Hashing Method
Compression Task	uses	L1 Regularized Logistic Regression Method
Compression Task	uses	Randomized Quantization Method
Noun Phrase Chunking Task	benefitsFrom	Compression Task
Small Sized Models OtherEntity	requires	Compression Task