Differentiable product quantization compresses embedding layers by replacing them with learned codebooks, achieving 14-238x compression with negligible performance loss on language tasks.
Proof of Proposition 1 Proof
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Differentiable Product Quantization for End-to-End Embedding Compression
Differentiable product quantization compresses embedding layers by replacing them with learned codebooks, achieving 14-238x compression with negligible performance loss on language tasks.