Densifying Assumed-Sparse Tensors - Improving Memory Efficiency and MPI Collective Performance During Tensor Accumulation for Parallelized Training of Neural Machine Translation Models.
Derya CavdarValeriu CodreanuCan KarakusJohn A. Lockman IIIDamian PodareanuVikram A. SaletoreAlexander SergeevDon D. Smith IIVictor SuthichaiQuy TaSrinivas VaradharajanLucas A. WilsonRengan XuPei YangPublished in: ISC (2019)
Keyphrases
- machine translation
- high order
- tensor factorization
- cross lingual
- information extraction
- natural language processing
- statistical translation models
- finite state transducers
- language processing
- language independent
- statistical model
- target language
- natural language generation
- statistical models
- associative memory
- diffusion tensor
- cross language information retrieval
- chinese english
- natural language