Distributed Inference for Latent Dirichlet Allocation
David Newman, Arthur Asuncion, Padhraic Smyth, Max Welling

Poster ID T55

LDA: 1 year time = 1 year, memory = 400 GB

Distributed LDA: 1 day time = 1 day, memory < 1 GB/proc

MEDLINE/PubMed 8 million abstracts 700 million words

P1

P1

P2

P3

P1024

topics,  mixtures,