Online Learning for Latent Dirichlet Allocation. Hoffman, M., D., Blei, D., M., & Bach, F. Paper Website abstract bibtex We develop an online variational Bayes (VB) algorithm for Latent Dirichlet Al-location (LDA). Online LDA is based on online stochastic optimization with a natural gradient step, which we show converges to a local optimum of the VB objective function. It can handily analyze massive document collections, includ-ing those arriving in a stream. We study the performance of online LDA in several ways, including by fitting a 100-topic topic model to 3.3M articles from Wikipedia in a single pass. We demonstrate that online LDA finds topic models as good or better than those found with batch VB, and in a fraction of the time.
@article{
title = {Online Learning for Latent Dirichlet Allocation},
type = {article},
websites = {https://papers.nips.cc/paper/3902-online-learning-for-latent-dirichlet-allocation.pdf},
id = {ab7bae81-a341-3d9b-8ebf-12729cff1b08},
created = {2018-02-05T16:53:51.835Z},
accessed = {2018-02-05},
file_attached = {true},
profile_id = {371589bb-c770-37ff-8193-93c6f25ffeb1},
group_id = {f982cd63-7ceb-3aa2-ac7e-a953963d6716},
last_modified = {2018-02-05T16:53:55.841Z},
read = {false},
starred = {false},
authored = {false},
confirmed = {false},
hidden = {false},
private_publication = {false},
abstract = {We develop an online variational Bayes (VB) algorithm for Latent Dirichlet Al-location (LDA). Online LDA is based on online stochastic optimization with a natural gradient step, which we show converges to a local optimum of the VB objective function. It can handily analyze massive document collections, includ-ing those arriving in a stream. We study the performance of online LDA in several ways, including by fitting a 100-topic topic model to 3.3M articles from Wikipedia in a single pass. We demonstrate that online LDA finds topic models as good or better than those found with batch VB, and in a fraction of the time.},
bibtype = {article},
author = {Hoffman, Matthew D and Blei, David M and Bach, Francis}
}
Downloads: 0
{"_id":"bJCScsSkHwgWMTgcw","bibbaseid":"hoffman-blei-bach-onlinelearningforlatentdirichletallocation","downloads":0,"creationDate":"2018-02-07T16:22:57.275Z","title":"Online Learning for Latent Dirichlet Allocation","author_short":["Hoffman, M., D.","Blei, D., M.","Bach, F."],"year":null,"bibtype":"article","biburl":null,"bibdata":{"title":"Online Learning for Latent Dirichlet Allocation","type":"article","websites":"https://papers.nips.cc/paper/3902-online-learning-for-latent-dirichlet-allocation.pdf","id":"ab7bae81-a341-3d9b-8ebf-12729cff1b08","created":"2018-02-05T16:53:51.835Z","accessed":"2018-02-05","file_attached":"true","profile_id":"371589bb-c770-37ff-8193-93c6f25ffeb1","group_id":"f982cd63-7ceb-3aa2-ac7e-a953963d6716","last_modified":"2018-02-05T16:53:55.841Z","read":false,"starred":false,"authored":false,"confirmed":false,"hidden":false,"private_publication":false,"abstract":"We develop an online variational Bayes (VB) algorithm for Latent Dirichlet Al-location (LDA). Online LDA is based on online stochastic optimization with a natural gradient step, which we show converges to a local optimum of the VB objective function. It can handily analyze massive document collections, includ-ing those arriving in a stream. We study the performance of online LDA in several ways, including by fitting a 100-topic topic model to 3.3M articles from Wikipedia in a single pass. We demonstrate that online LDA finds topic models as good or better than those found with batch VB, and in a fraction of the time.","bibtype":"article","author":"Hoffman, Matthew D and Blei, David M and Bach, Francis","bibtex":"@article{\n title = {Online Learning for Latent Dirichlet Allocation},\n type = {article},\n websites = {https://papers.nips.cc/paper/3902-online-learning-for-latent-dirichlet-allocation.pdf},\n id = {ab7bae81-a341-3d9b-8ebf-12729cff1b08},\n created = {2018-02-05T16:53:51.835Z},\n accessed = {2018-02-05},\n file_attached = {true},\n profile_id = {371589bb-c770-37ff-8193-93c6f25ffeb1},\n group_id = {f982cd63-7ceb-3aa2-ac7e-a953963d6716},\n last_modified = {2018-02-05T16:53:55.841Z},\n read = {false},\n starred = {false},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {We develop an online variational Bayes (VB) algorithm for Latent Dirichlet Al-location (LDA). Online LDA is based on online stochastic optimization with a natural gradient step, which we show converges to a local optimum of the VB objective function. It can handily analyze massive document collections, includ-ing those arriving in a stream. We study the performance of online LDA in several ways, including by fitting a 100-topic topic model to 3.3M articles from Wikipedia in a single pass. We demonstrate that online LDA finds topic models as good or better than those found with batch VB, and in a fraction of the time.},\n bibtype = {article},\n author = {Hoffman, Matthew D and Blei, David M and Bach, Francis}\n}","author_short":["Hoffman, M., D.","Blei, D., M.","Bach, F."],"urls":{"Paper":"http://bibbase.org/service/mendeley/371589bb-c770-37ff-8193-93c6f25ffeb1/file/57adc406-4bf4-b865-0e5f-f0f060fca863/Online_Learning_for_Latent_Dirichlet_Allocation.pdf.pdf","Website":"https://papers.nips.cc/paper/3902-online-learning-for-latent-dirichlet-allocation.pdf"},"bibbaseid":"hoffman-blei-bach-onlinelearningforlatentdirichletallocation","role":"author","downloads":0,"html":""},"search_terms":["online","learning","latent","dirichlet","allocation","hoffman","blei","bach"],"keywords":[],"authorIDs":[]}