Advances in Variational Inference
Many modern unsupervised or semi-supervised machine learning algorithms rely on Bayesian probabilistic models. These models are usually intractable and thus require approximate inference. Variational inference (VI) lets us approximate a high-dimensional Bayesian posterior with a simpler variational...
Saved in:
| Published in: | IEEE transactions on pattern analysis and machine intelligence Vol. 41; no. 8; pp. 2008 - 2026 |
|---|---|
| Main Authors: | , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
United States
IEEE
01.08.2019
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Subjects: | |
| ISSN: | 0162-8828, 1939-3539, 2160-9292, 1939-3539 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Many modern unsupervised or semi-supervised machine learning algorithms rely on Bayesian probabilistic models. These models are usually intractable and thus require approximate inference. Variational inference (VI) lets us approximate a high-dimensional Bayesian posterior with a simpler variational distribution by solving an optimization problem. This approach has been successfully applied to various models and large-scale applications. In this review, we give an overview of recent trends in variational inference. We first introduce standard mean field variational inference, then review recent advances focusing on the following aspects: (a) scalable VI, which includes stochastic approximations, (b) generic VI, which extends the applicability of VI to a large class of otherwise intractable models, such as non-conjugate models, (c) accurate VI, which includes variational models beyond the mean field approximation or with atypical divergences, and (d) amortized VI, which implements the inference over local latent variables with inference networks. Finally, we provide a summary of promising future research directions. |
|---|---|
| AbstractList | Many modern unsupervised or semi-supervised machine learning algorithms rely on Bayesian probabilistic models. These models are usually intractable and thus require approximate inference. Variational inference (VI) lets us approximate a high-dimensional Bayesian posterior with a simpler variational distribution by solving an optimization problem. This approach has been successfully applied to various models and large-scale applications. In this review, we give an overview of recent trends in variational inference. We first introduce standard mean field variational inference, then review recent advances focusing on the following aspects: (a) scalable VI, which includes stochastic approximations, (b) generic VI, which extends the applicability of VI to a large class of otherwise intractable models, such as non-conjugate models, (c) accurate VI, which includes variational models beyond the mean field approximation or with atypical divergences, and (d) amortized VI, which implements the inference over local latent variables with inference networks. Finally, we provide a summary of promising future research directions.Many modern unsupervised or semi-supervised machine learning algorithms rely on Bayesian probabilistic models. These models are usually intractable and thus require approximate inference. Variational inference (VI) lets us approximate a high-dimensional Bayesian posterior with a simpler variational distribution by solving an optimization problem. This approach has been successfully applied to various models and large-scale applications. In this review, we give an overview of recent trends in variational inference. We first introduce standard mean field variational inference, then review recent advances focusing on the following aspects: (a) scalable VI, which includes stochastic approximations, (b) generic VI, which extends the applicability of VI to a large class of otherwise intractable models, such as non-conjugate models, (c) accurate VI, which includes variational models beyond the mean field approximation or with atypical divergences, and (d) amortized VI, which implements the inference over local latent variables with inference networks. Finally, we provide a summary of promising future research directions. Many modern unsupervised or semi-supervised machine learning algorithms rely on Bayesian probabilistic models. These models are usually intractable and thus require approximate inference. Variational inference (VI) lets us approximate a high-dimensional Bayesian posterior with a simpler variational distribution by solving an optimization problem. This approach has been successfully applied to various models and large-scale applications. In this review, we give an overview of recent trends in variational inference. We first introduce standard mean field variational inference, then review recent advances focusing on the following aspects: (a) scalable VI, which includes stochastic approximations, (b) generic VI, which extends the applicability of VI to a large class of otherwise intractable models, such as non-conjugate models, (c) accurate VI, which includes variational models beyond the mean field approximation or with atypical divergences, and (d) amortized VI, which implements the inference over local latent variables with inference networks. Finally, we provide a summary of promising future research directions. Many modern unsupervised or semi-supervised machine learning algorithms rely on Bayesian probabilistic models. These models are usually intractable and thus require approximate inference. Variational inference (VI) lets us approximate a high-dimensional Bayesian posterior with a simpler variational distribution by solving an optimization problem. This approach has been successfully applied to various models and large-scale applications. In this review, we give an overview of recent trends in variational inference. We first introduce standard mean field variational inference, then review recent advances focusing on the following aspects: (a) scalable VI, which includes stochastic approximations, (b) generic VI, which extends the applicability of VI to a large class of otherwise intractable models, such as non-conjugate models, mean field approximation or with atypical divergences, and (d) amortized VI, which implements the inference over local latent variables with inference networks. Finally, we provide a summary of promising future research directions. |
| Author | Butepage, Judith Mandt, Stephan Zhang, Cheng Kjellstrom, Hedvig |
| Author_xml | – sequence: 1 givenname: Cheng orcidid: 0000-0002-8640-9370 surname: Zhang fullname: Zhang, Cheng email: cheng.zhang@microsoft.com organization: Microsoft Research, Cambridge, United Kingdom – sequence: 2 givenname: Judith orcidid: 0000-0001-5344-8042 surname: Butepage fullname: Butepage, Judith email: butepage@kth.se organization: KTH Royal Institute of Technology, Stockholm, Sweden – sequence: 3 givenname: Hedvig surname: Kjellstrom fullname: Kjellstrom, Hedvig email: hedvig@kth.se organization: KTH Royal Institute of Technology, Stockholm, Sweden – sequence: 4 givenname: Stephan surname: Mandt fullname: Mandt, Stephan email: stephan.mandt@gmail.com organization: University of California, Irvine, Irvine, CA, USA |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/30596568$$D View this record in MEDLINE/PubMed https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-255405$$DView record from Swedish Publication Index (Kungliga Tekniska Högskolan) |
| BookMark | eNp9kTtPwzAUhS0EgvL4AyChIhaWFD9je6x4VgLBULparnMDhjQpdgLi3-PS0oGB6Q73O_dxzi7arJsaEDokeEAI1ufjx-H9aEAxUQOqlJaSb6AeJTnONNV0E_UwyWmmFFU7aDfGV4wJF5htox2Ghc5FrnroZFh82NpB7Pu6P7HB29Y3ta36o7qEAKmzj7ZKW0U4WNU99HR9Nb64ze4ebkYXw7vMccbajOiyYLKUFrSe8pKAAMplydJ1hBHpWM6EZADE2TyHYkqZLDhhrlSWOweM7aFsOTd-wrybmnnwMxu-TGO9ufSToWnCs3lrXwwVgmOR-LMlPw_NewexNTMfHVSVraHpoklGUK45Jwv09A_62nQhfZkoKrAkC48SdbyiuukMivUBv2YlgC4BF5oYA5RrhGCzSMT8JGIWiZhVIkmk_oicb39MboP11f_So6XUA8B6lxJKMa3ZN6ZDlfc |
| CODEN | ITPIDJ |
| CitedBy_id | crossref_primary_10_1145_3451528 crossref_primary_10_1109_TPAMI_2022_3185773 crossref_primary_10_3390_e23111475 crossref_primary_10_1007_s10994_022_06169_w crossref_primary_10_3390_e22111263 crossref_primary_10_1007_s10182_024_00492_4 crossref_primary_10_1007_s41060_024_00580_3 crossref_primary_10_1109_ACCESS_2024_3466919 crossref_primary_10_1109_TSG_2019_2931160 crossref_primary_10_1371_journal_pone_0247795 crossref_primary_10_3390_math11214486 crossref_primary_10_1016_j_actamat_2024_119877 crossref_primary_10_3390_act12020070 crossref_primary_10_1016_j_jlp_2025_105768 crossref_primary_10_1016_j_energy_2023_127286 crossref_primary_10_1109_TSG_2020_2998080 crossref_primary_10_1088_1361_6420_ab15a3 crossref_primary_10_1109_JIOT_2025_3545739 crossref_primary_10_1016_j_ajhg_2023_03_009 crossref_primary_10_1109_TMI_2024_3364911 crossref_primary_10_1007_s44379_024_00011_x crossref_primary_10_1016_j_sigpro_2025_109954 crossref_primary_10_1109_TIM_2023_3265101 crossref_primary_10_1109_TIP_2023_3295929 crossref_primary_10_1145_3371087 crossref_primary_10_1214_22_STS876 crossref_primary_10_1017_S0140525X21002351 crossref_primary_10_3390_e25081235 crossref_primary_10_1016_j_radphyschem_2024_112389 crossref_primary_10_1109_TNNLS_2021_3054407 crossref_primary_10_1109_TII_2022_3158615 crossref_primary_10_3390_info15060336 crossref_primary_10_1029_2022JB024098 crossref_primary_10_1080_03610926_2021_1921214 crossref_primary_10_1016_j_istruc_2025_109170 crossref_primary_10_1109_TSP_2023_3327845 crossref_primary_10_1007_s00477_025_02954_w crossref_primary_10_1007_s42985_021_00102_x crossref_primary_10_1109_TRS_2025_3586927 crossref_primary_10_1093_bib_bbaf283 crossref_primary_10_1016_j_ejor_2024_08_024 crossref_primary_10_1109_TWC_2024_3371703 crossref_primary_10_1214_22_STS875 crossref_primary_10_1016_j_rser_2021_110930 crossref_primary_10_1007_s11222_022_10172_5 crossref_primary_10_1137_19M130409X crossref_primary_10_1016_j_sigpro_2022_108571 crossref_primary_10_1017_asb_2024_7 crossref_primary_10_3389_fenrg_2023_1162068 crossref_primary_10_1016_j_jcp_2022_111454 crossref_primary_10_1007_s11222_024_10531_4 crossref_primary_10_1016_j_istruc_2025_109063 crossref_primary_10_1016_j_apacoust_2023_109512 crossref_primary_10_1109_LSP_2021_3099972 crossref_primary_10_1007_s10479_025_06693_7 crossref_primary_10_1016_j_cmpb_2024_108060 crossref_primary_10_1007_s41060_021_00290_0 crossref_primary_10_1016_j_dsp_2024_104627 crossref_primary_10_1016_j_patcog_2023_110113 crossref_primary_10_1016_j_jpdc_2024_104916 crossref_primary_10_1016_j_sigpro_2023_108966 crossref_primary_10_1016_j_ijggc_2025_104433 crossref_primary_10_1007_s11134_022_09830_2 crossref_primary_10_1016_j_ins_2022_01_051 crossref_primary_10_1038_s41598_021_00144_6 crossref_primary_10_1007_s11135_022_01460_3 crossref_primary_10_1016_j_aca_2025_344308 crossref_primary_10_1039_D4AY02222J crossref_primary_10_1007_s11222_023_10339_8 crossref_primary_10_1016_j_knosys_2023_111315 crossref_primary_10_1016_j_patcog_2025_112442 crossref_primary_10_1137_21M1440773 crossref_primary_10_1109_TSMC_2023_3312585 crossref_primary_10_1007_s11222_025_10654_2 crossref_primary_10_1016_j_ijforecast_2024_05_005 crossref_primary_10_1214_25_BA1542 crossref_primary_10_1109_TGRS_2023_3257865 crossref_primary_10_1186_s40537_024_01005_5 crossref_primary_10_1016_j_ymssp_2023_110796 crossref_primary_10_1007_s11336_022_09884_4 crossref_primary_10_1016_j_conbuildmat_2024_135466 crossref_primary_10_1145_3715876 crossref_primary_10_1007_s10928_024_09931_w crossref_primary_10_1016_j_ymssp_2024_112113 crossref_primary_10_1371_journal_pcbi_1012608 crossref_primary_10_1109_TUFFC_2024_3466290 crossref_primary_10_1016_j_engappai_2021_104423 crossref_primary_10_1016_j_sigpro_2022_108826 crossref_primary_10_1109_TVT_2025_3534820 crossref_primary_10_1371_journal_pcbi_1011395 crossref_primary_10_1016_j_ecoinf_2024_102853 crossref_primary_10_1016_j_sigpro_2024_109478 crossref_primary_10_1080_10618600_2025_2510494 crossref_primary_10_1016_j_neucom_2023_126960 crossref_primary_10_1016_j_automatica_2021_109827 crossref_primary_10_1016_j_rineng_2025_105267 crossref_primary_10_1080_01621459_2023_2296704 crossref_primary_10_1002_pc_29998 crossref_primary_10_1109_TIP_2023_3326687 crossref_primary_10_1109_TPAMI_2022_3153225 crossref_primary_10_1109_TGRS_2024_3394462 crossref_primary_10_1016_j_ymssp_2024_112261 crossref_primary_10_1109_TNNLS_2023_3245329 crossref_primary_10_1145_3386252 crossref_primary_10_1016_j_trb_2022_01_005 crossref_primary_10_1016_j_eswa_2022_119380 crossref_primary_10_3390_pr13030861 crossref_primary_10_1007_s00180_023_01350_8 crossref_primary_10_1109_TIM_2022_3192054 crossref_primary_10_1016_j_ymssp_2024_112253 crossref_primary_10_1111_str_70008 crossref_primary_10_1007_s10489_021_03100_z crossref_primary_10_3390_e23010117 crossref_primary_10_1109_TPAMI_2022_3140886 crossref_primary_10_1016_j_jfranklin_2025_107548 crossref_primary_10_1109_ACCESS_2023_3247947 crossref_primary_10_1016_j_compstruct_2024_118087 crossref_primary_10_1016_j_trc_2020_102622 crossref_primary_10_1002_rnc_5916 crossref_primary_10_1007_s10915_023_02328_w crossref_primary_10_1098_rsif_2024_0880 crossref_primary_10_1109_TPAMI_2020_3001433 crossref_primary_10_1016_j_ress_2023_109547 crossref_primary_10_1098_rsta_2024_0324 crossref_primary_10_1007_s11222_022_10189_w crossref_primary_10_1109_TKDE_2023_3327415 crossref_primary_10_1088_2634_4386_aca7de crossref_primary_10_1109_JPROC_2023_3328920 crossref_primary_10_1109_TMM_2023_3268870 crossref_primary_10_1088_1361_6501_ad5905 crossref_primary_10_1016_j_net_2025_103679 crossref_primary_10_3390_e26020129 crossref_primary_10_1109_TSP_2025_3540018 crossref_primary_10_1080_10618600_2024_2402278 crossref_primary_10_1016_j_cma_2024_117359 crossref_primary_10_1109_TAES_2022_3184283 crossref_primary_10_1088_1475_7516_2023_07_065 crossref_primary_10_3390_electronics14101908 crossref_primary_10_1371_journal_pcbi_1011280 crossref_primary_10_26443_seismica_v3i1_1143 crossref_primary_10_1080_10618600_2022_2130927 crossref_primary_10_1121_10_0017064 crossref_primary_10_1007_s00158_023_03610_z crossref_primary_10_1088_2632_2153_ace2aa crossref_primary_10_1002_aic_17753 crossref_primary_10_1016_j_cma_2022_114712 crossref_primary_10_1007_s40435_025_01818_z crossref_primary_10_1016_j_automatica_2024_112086 crossref_primary_10_1088_1742_5468_ab43d3 crossref_primary_10_1016_j_biopsycho_2024_108891 crossref_primary_10_1016_j_electacta_2024_144694 crossref_primary_10_1214_20_AOS2035 crossref_primary_10_1109_TSP_2022_3164200 crossref_primary_10_3389_frobt_2020_00047 crossref_primary_10_1109_TPAMI_2023_3316671 crossref_primary_10_1109_JSTSP_2021_3058019 crossref_primary_10_1190_geo2023_0035_1 crossref_primary_10_1016_j_sigpro_2021_108226 crossref_primary_10_1137_24M1642202 crossref_primary_10_1109_TNNLS_2023_3300977 crossref_primary_10_1016_j_engstruct_2025_119709 crossref_primary_10_1093_gji_ggaf239 crossref_primary_10_1016_j_adhoc_2025_103887 crossref_primary_10_1093_jrsssc_qlad029 crossref_primary_10_1109_JIOT_2021_3113679 crossref_primary_10_1016_j_media_2024_103228 crossref_primary_10_1109_ACCESS_2024_3452476 crossref_primary_10_1109_TII_2024_3507940 crossref_primary_10_1088_2632_2153_ad9a39 crossref_primary_10_1016_j_automatica_2023_111314 crossref_primary_10_1109_TWC_2023_3268463 crossref_primary_10_1016_j_neuroimage_2022_119595 crossref_primary_10_1090_mcom_4013 crossref_primary_10_1109_LRA_2021_3137503 crossref_primary_10_1186_s13059_020_02015_1 crossref_primary_10_4218_etrij_2024_0491 crossref_primary_10_3390_s20205966 crossref_primary_10_1109_LSP_2021_3113279 crossref_primary_10_1109_TCSVT_2023_3311039 crossref_primary_10_1109_TAES_2024_3491936 crossref_primary_10_3934_math_2025977 crossref_primary_10_1016_j_renene_2025_122892 crossref_primary_10_1007_s10489_024_06125_2 crossref_primary_10_1016_j_ijforecast_2023_05_002 crossref_primary_10_1109_TCCN_2024_3457525 crossref_primary_10_1007_s10712_024_09847_7 crossref_primary_10_3390_informatics12020038 crossref_primary_10_1016_j_jcp_2024_113117 crossref_primary_10_1109_ACCESS_2025_3533878 crossref_primary_10_1016_j_ipm_2025_104336 crossref_primary_10_1002_sim_10072 crossref_primary_10_3389_fninf_2024_1392661 crossref_primary_10_1109_TAI_2022_3215614 crossref_primary_10_1109_TMC_2024_3362912 crossref_primary_10_1080_10618600_2022_2134875 crossref_primary_10_1080_10618600_2021_1923516 crossref_primary_10_1109_TMC_2023_3325301 crossref_primary_10_1109_TSP_2021_3122296 crossref_primary_10_1016_j_trc_2025_105215 crossref_primary_10_1109_ACCESS_2021_3127448 crossref_primary_10_1016_j_knosys_2022_108827 crossref_primary_10_3390_en14144235 crossref_primary_10_1109_ACCESS_2020_3028185 crossref_primary_10_1016_j_knosys_2022_110043 crossref_primary_10_1145_3450963 crossref_primary_10_1145_3522762 crossref_primary_10_1088_1361_6501_ad27c6 crossref_primary_10_1016_j_patcog_2019_107156 crossref_primary_10_1016_j_neucom_2022_07_019 crossref_primary_10_1109_ACCESS_2020_2981948 crossref_primary_10_1016_j_jii_2024_100728 crossref_primary_10_1109_TNNLS_2022_3213315 crossref_primary_10_1039_D1SC07210B crossref_primary_10_3847_1538_4357_ac6be9 crossref_primary_10_1109_TPAMI_2021_3133763 crossref_primary_10_12688_wellcomeopenres_15881_2 crossref_primary_10_1016_j_neunet_2024_106246 crossref_primary_10_12688_wellcomeopenres_15881_1 crossref_primary_10_1038_s41598_022_10966_7 crossref_primary_10_1016_j_automatica_2022_110374 crossref_primary_10_1016_j_patrec_2022_10_024 crossref_primary_10_1109_TIT_2022_3229765 crossref_primary_10_1109_TRO_2019_2959445 crossref_primary_10_1109_TAES_2023_3323629 crossref_primary_10_1038_s41467_025_60269_4 crossref_primary_10_1109_ACCESS_2022_3187839 crossref_primary_10_1137_19M1267246 crossref_primary_10_1109_TAES_2024_3392552 crossref_primary_10_1109_TNNLS_2021_3105570 crossref_primary_10_1109_TITS_2024_3454768 crossref_primary_10_1109_TIP_2022_3183835 crossref_primary_10_1090_mcom_3906 crossref_primary_10_1109_TSMC_2025_3564984 crossref_primary_10_3390_math12213319 crossref_primary_10_1016_j_epsr_2020_106253 crossref_primary_10_1080_10618600_2024_2374962 crossref_primary_10_1109_ACCESS_2024_3510927 crossref_primary_10_1016_j_trc_2022_103571 crossref_primary_10_1109_TAES_2025_3559523 crossref_primary_10_1109_TBME_2021_3103141 crossref_primary_10_1007_s11634_023_00569_z crossref_primary_10_1016_j_rineng_2024_102773 crossref_primary_10_1109_TCE_2024_3507006 crossref_primary_10_1007_s11222_024_10488_4 crossref_primary_10_3390_robotics11020038 crossref_primary_10_1093_jrsssc_qlad102 crossref_primary_10_1016_j_sigpro_2021_108314 crossref_primary_10_1109_ACCESS_2023_3237990 crossref_primary_10_1109_TNNLS_2022_3149285 crossref_primary_10_1016_j_neucom_2024_128910 crossref_primary_10_1007_s41237_022_00164_0 crossref_primary_10_1145_3696412 crossref_primary_10_1002_wics_70005 crossref_primary_10_1016_j_ress_2025_110833 crossref_primary_10_3847_1538_4357_ac7a3c crossref_primary_10_1109_ACCESS_2023_3335196 crossref_primary_10_1007_s11071_021_06626_6 crossref_primary_10_3390_math10060898 crossref_primary_10_1016_j_patcog_2023_110037 crossref_primary_10_1038_s44296_024_00011_1 crossref_primary_10_1145_3571243 crossref_primary_10_3847_1538_4357_ad063f crossref_primary_10_3390_machines13030216 crossref_primary_10_1109_TPAMI_2019_2950025 crossref_primary_10_1109_TPAMI_2023_3243306 crossref_primary_10_1080_24725854_2024_2302368 crossref_primary_10_1007_s11336_021_09748_3 crossref_primary_10_1049_gtd2_12603 crossref_primary_10_1016_j_csda_2020_107012 crossref_primary_10_1016_j_neuroimage_2020_117366 crossref_primary_10_1109_TIM_2023_3291801 crossref_primary_10_1111_bmsp_12363 crossref_primary_10_1137_23M1618922 crossref_primary_10_1109_MSP_2022_3198201 crossref_primary_10_1007_s10994_024_06513_2 crossref_primary_10_1016_j_chemolab_2025_105378 crossref_primary_10_1140_epjds_s13688_025_00544_y crossref_primary_10_1007_s12065_023_00900_9 crossref_primary_10_1016_j_cma_2024_117195 crossref_primary_10_1080_01621459_2025_2498088 crossref_primary_10_1016_j_phycom_2025_102692 crossref_primary_10_1007_s00477_025_02956_8 crossref_primary_10_1007_s10462_024_10859_3 crossref_primary_10_3390_e25101468 crossref_primary_10_1016_j_jobe_2021_102830 crossref_primary_10_1145_3477140 crossref_primary_10_1007_s11804_024_00415_4 crossref_primary_10_1016_j_neucom_2024_128934 crossref_primary_10_1093_bib_bbae689 crossref_primary_10_3389_fnbeh_2021_633872 crossref_primary_10_1016_j_chemolab_2024_105088 crossref_primary_10_1360_SSM_2024_0100 crossref_primary_10_1007_s10618_022_00849_w crossref_primary_10_1109_TSP_2020_2975353 crossref_primary_10_1109_TCOMM_2025_3541076 crossref_primary_10_1007_s12046_022_01980_6 crossref_primary_10_1109_TPAMI_2023_3329339 |
| Cites_doi | 10.1214/aoms/1177729586 10.1201/b10905 10.1103/PhysRevLett.86.3695 10.1023/A:1007665907178 10.1613/jair.251 10.1162/089976698300017746 10.1214/aoms/1177729694 10.1080/01621459.2017.1285773 10.1214/13-BA858 10.1007/978-3-7908-2604-3_16 10.1162/089976601750265045 10.1145/1143844.1143859 10.1145/1401890.1401960 10.1109/TIT.2005.850091 10.1109/CVPR.2017.637 10.1007/978-1-4612-5056-2 10.1016/j.patcog.2011.10.004 10.1201/9781315366951 10.1080/14786437708235992 10.1016/0304-405X(77)90005-8 10.1007/s11222-015-9618-x 10.1023/A:1022672621406 10.1561/2200000052 10.1214/ss/1177013621 10.1016/S0959-440X(96)80056-X 10.1145/2187836.2187955 10.1111/j.2517-6161.1966.tb00626.x 10.2307/2287970 10.1162/089976600300015213 10.1162/neco.1995.7.5.889 10.1109/ICDMW.2007.33 10.1111/rssb.12185 10.1016/0167-8655(90)90010-Y 10.2307/1428011 10.1007/978-94-011-5014-9_6 10.1162/neco.2007.19.4.1112 10.1109/ICCVW.2013.41 10.1007/s10107-012-0572-5 10.1088/0305-4470/15/6/035 10.18653/v1/K16-1002 10.1109/TIT.2009.2030485 10.1561/2200000001 10.1137/110830629 10.7551/mitpress/1100.001.0001 10.1145/3064899.3064910 10.1109/TPAMI.2008.157 10.1038/nature14541 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2019 |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2019 |
| DBID | 97E ESBDL RIA RIE AAYXX CITATION NPM 7SC 7SP 8FD JQ2 L7M L~C L~D 7X8 ADTPV AFDQA AOWAS D8T D8V ZZAVC |
| DOI | 10.1109/TPAMI.2018.2889774 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE Xplore Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef PubMed Computer and Information Systems Abstracts Electronics & Communications Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional MEDLINE - Academic SwePub SWEPUB Kungliga Tekniska Högskolan full text SwePub Articles SWEPUB Freely available online SWEPUB Kungliga Tekniska Högskolan SwePub Articles full text |
| DatabaseTitle | CrossRef PubMed Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic Technology Research Database PubMed |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher – sequence: 3 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering Computer Science |
| EISSN | 2160-9292 1939-3539 |
| EndPage | 2026 |
| ExternalDocumentID | oai_DiVA_org_kth_255405 30596568 10_1109_TPAMI_2018_2889774 8588399 |
| Genre | orig-research Journal Article |
| GroupedDBID | --- -DZ -~X .DC 0R~ 29I 4.4 53G 5GY 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACGFS ACIWK ACNCT AENEX AGQYO AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 E.L EBS EJD ESBDL F5P HZ~ IEDLZ IFIPE IPLJI JAVBF LAI M43 MS~ O9- OCL P2P PQQKQ RIA RIE RNS RXW TAE TN5 UHB ~02 AAYXX CITATION 5VS 9M8 ABFSI ADRHT AETEA AETIX AGSQL AI. AIBXA ALLEH FA8 H~9 IBMZZ ICLAB IFJZH NPM RIG RNI RZB VH1 XJT 7SC 7SP 8FD JQ2 L7M L~C L~D 7X8 ADTPV AFDQA AOWAS D8T D8V ZZAVC |
| ID | FETCH-LOGICAL-c433t-19fd37f7ae99b4f1e5e247f30181317c363573ee1ca66edb237d413cf8a4cce33 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 428 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000473598800016&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0162-8828 1939-3539 |
| IngestDate | Tue Nov 04 16:42:57 EST 2025 Sat Sep 27 16:46:44 EDT 2025 Mon Jun 30 07:17:18 EDT 2025 Mon Jul 21 05:58:20 EDT 2025 Sat Nov 29 05:15:58 EST 2025 Tue Nov 18 21:45:11 EST 2025 Wed Aug 27 05:51:41 EDT 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 8 |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/OAPA.html |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c433t-19fd37f7ae99b4f1e5e247f30181317c363573ee1ca66edb237d413cf8a4cce33 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| ORCID | 0000-0002-8640-9370 0000-0001-5344-8042 |
| OpenAccessLink | https://ieeexplore.ieee.org/document/8588399 |
| PMID | 30596568 |
| PQID | 2250718828 |
| PQPubID | 85458 |
| PageCount | 19 |
| ParticipantIDs | proquest_journals_2250718828 crossref_primary_10_1109_TPAMI_2018_2889774 ieee_primary_8588399 proquest_miscellaneous_2162494415 pubmed_primary_30596568 swepub_primary_oai_DiVA_org_kth_255405 crossref_citationtrail_10_1109_TPAMI_2018_2889774 |
| PublicationCentury | 2000 |
| PublicationDate | 2019-08-01 |
| PublicationDateYYYYMMDD | 2019-08-01 |
| PublicationDate_xml | – month: 08 year: 2019 text: 2019-08-01 day: 01 |
| PublicationDecade | 2010 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States – name: New York |
| PublicationTitle | IEEE transactions on pattern analysis and machine intelligence |
| PublicationTitleAbbrev | TPAMI |
| PublicationTitleAlternate | IEEE Trans Pattern Anal Mach Intell |
| PublicationYear | 2019 |
| Publisher | IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Publisher_xml | – name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| References | ref204 jang (ref75) 2017 broderick (ref21) 2013 wang (ref209) 2011 mescheder (ref119) 2017 de (ref37) 2017 karaletsos (ref81) 2016 goodfellow (ref53) 2016 ref46 titsias (ref196) 0 ref219 wang (ref211) 2017 ref215 peterson (ref149) 1987; 1 ref44 tucker (ref203) 2017 ranganath (ref155) 2016 seeger (ref174) 2005 ref7 alemi (ref2) 2017 (ref148) 2017 ref3 snelson (ref177) 2006 ref6 ref5 ref222 johnson (ref78) 2013 titsias (ref197) 2011 mandt (ref115) 2016 rolfe (ref163) 2017 wang (ref210) 2016 ref36 dinh (ref41) 2017 foti (ref45) 2014 kurihara (ref94) 2007 csiba (ref32) 2018; 19 maddison (ref111) 2017 ref38 minka (ref126) 2014 schulman (ref173) 2015 murphy (ref131) 1999 zeiler (ref218) 2012 gal (ref49) 2014 minka (ref125) 2004 krishnan (ref91) 2015 ref26 honkela (ref69) 2010; 11 carpenter (ref28) 2016; 76 ref20 johnson (ref77) 2016 huszár (ref72) 2017 hensman (ref64) 2012 hensman (ref63) 2013 opper (ref144) 1996 le cam (ref27) 0 ref97 ranganath (ref157) 2013 ref96 tanaka (ref186) 1999 kucukelbir (ref92) 2015 wang (ref212) 2017 sutton (ref183) 1998; 1 roeder (ref162) 2017 liu (ref109) 2016 ref93 ref133 opper (ref142) 2013; 14 ruiz (ref167) 2016 duvenaud (ref43) 2016 ref130 li (ref100) 2018 bishop (ref14) 2006 honkela (ref70) 2007 ref135 han (ref60) 2016 wainwright (ref205) 2008; 1 raymond (ref158) 2014 paisley (ref146) 2012 damianou (ref35) 2013 zhao (ref227) 2017 ref145 ref143 ref140 tran (ref201) 2016; 1050 cremer (ref30) 2018 amari (ref4) 1985 ref79 brooks (ref22) 2011 teh (ref188) 2006 ref74 tran (ref200) 2016 theis (ref189) 2015 csiszár (ref33) 1964; 8 mescheder (ref118) 2017 rainforth (ref152) 2018 zhao (ref225) 2014 li (ref102) 2016 miao (ref122) 2016 bui (ref24) 2016; 18 ranganath (ref156) 2016 zhao (ref226) 2015 tieleman (ref191) 2012; 4 ref110 naesseth (ref132) 2017 king (ref83) 2006 mandt (ref113) 2016 sriperumbudur (ref179) 2009 gulrajani (ref57) 2017 liu (ref107) 2016 rainforth (ref153) 2018 yin (ref217) 2018 wang (ref208) 2013 gopalan (ref55) 2012 marino (ref116) 2018 li (ref101) 2015 wang (ref213) 2018 zhu (ref229) 1995 neal (ref136) 1993 ref170 opper (ref141) 2015 titsias (ref195) 2009 honkela (ref71) 2003 ref171 ref172 mandt (ref114) 2017; 18 saeedi (ref168) 0 gregor (ref56) 2015 dai (ref34) 2016 korattikara (ref90) 2015 balles (ref10) 2017 nowozin (ref139) 2018 liu (ref108) 2016 bamler (ref12) 2017 mandt (ref112) 2014 srivastava (ref180) 2017 ruiz (ref166) 2016 barber (ref13) 2003 ref187 ref184 ref182 gershman (ref51) 2012 salimans (ref169) 2015 hoffman (ref65) 2010 kingma (ref84) 2016 hogg (ref68) 1995 wang (ref206) 2009 johnson (ref76) 2014 tran (ref202) 0 blei (ref16) 2006 li (ref103) 2016 ghahramani (ref52) 2015; 521 levine (ref98) 0 ref151 bamler (ref11) 2017 ref150 dinh (ref40) 2015 cremer (ref31) 2017 hennig (ref62) 2011 welling (ref214) 2011 ahn (ref1) 2012 gershman (ref50) 2014; 36 gal (ref48) 2016 zhang (ref223) 2017 tanaka (ref185) 1998 ranganath (ref154) 2014 ref164 neiswanger (ref138) 2015 li (ref104) 2016 ref161 khan (ref82) 2015 ref15 azevedo-filho and (ref9) 1994 sønderby (ref178) 2016 ref17 ref19 ref18 burda (ref25) 2016 kingma (ref85) 2014 tishby (ref193) 2000 hoffman (ref67) 2013; 14 jaakkol (ref73) 1996 levine (ref99) 2013 lin (ref106) 2018 rezende (ref159) 2015 hoffman (ref66) 2015 zhang (ref221) 2017 neal (ref137) 2012; 118 li (ref105) 2017 parisi (ref147) 1988 rezende (ref160) 2014 minka (ref127) 2001 han (ref59) 2017 shah (ref175) 2015 knowles (ref89) 2015 stein (ref181) 1972 dieng (ref39) 2017 nalisnick (ref134) 2017 winn (ref216) 2005; 6 arenz (ref8) 2018 zhang (ref220) 2016 knowles (ref88) 2011 miller (ref123) 2017 buchholz (ref23) 2018 kappen (ref80) 2001 kingma (ref87) 2015 minka (ref128) 2005 duchi (ref42) 2011; 12 titsias (ref194) 2015 mohamed (ref129) 2016 mcinerney (ref117) 2015 khan (ref121) 0 ref192 mézard (ref120) 1990; 9 ref190 wang (ref207) 2013; 14 zhang (ref224) 0 ref198 fu (ref47) 2017 lacoste-julien (ref95) 2011 guo (ref58) 2016 shi (ref176) 2017 tran (ref199) 2015 zhe (ref228) 2016 chen (ref29) 2017 ross (ref165) 2006 heinrich (ref61) 2008 goodfellow (ref54) 2014 miller (ref124) 2017 kingma (ref86) 2014 |
| References_xml | – start-page: 354 year: 2016 ident: ref113 article-title: A variational analysis of stochastic gradient algorithms publication-title: Proc 33rd Int Conf Mach Learn – ident: ref161 doi: 10.1214/aoms/1177729586 – start-page: 1727 year: 2013 ident: ref21 article-title: Streaming variational bayes publication-title: Proc 26th Int Conf Neural Inf Process Syst – start-page: 123 year: 2012 ident: ref1 article-title: Bayesian posterior sampling via stochastic gradient fisher scoring publication-title: Proc 29th Int Conf Mach Learn – start-page: 1511 year: 2016 ident: ref104 article-title: Black-box alpha divergence minimization publication-title: Proc 33rd Int Conf Mach Learn – volume: 118 year: 2012 ident: ref137 publication-title: Bayesian learning for neural networks – start-page: 362 year: 2001 ident: ref127 article-title: Expectation propagation for approximate Bayesian inference publication-title: Proc 17th Int Conf Uncertainty Artif Intell – start-page: 1 year: 2015 ident: ref226 article-title: Stochastic optimization with importance sampling for regularized loss minimization publication-title: Proceedings of the 32nd Intl Conf on Machine Learning – year: 2016 ident: ref107 article-title: Two methods for wild variational inference publication-title: arXiv 1612 00081 – year: 2011 ident: ref22 publication-title: Handbook of Markov Chain Monte Carlo doi: 10.1201/b10905 – start-page: 814 year: 2014 ident: ref154 article-title: Black box variational inference publication-title: Proc 17th Int Conf Artif Intell Statist – ident: ref145 doi: 10.1103/PhysRevLett.86.3695 – start-page: ii-298 year: 2013 ident: ref157 article-title: An adaptive learning rate for stochastic variational inference publication-title: Proc 30th Int Conf Mach Learn – start-page: 752 year: 2011 ident: ref209 article-title: Online variational inference for the hierarchical dirichlet process publication-title: Proc 14th Int Conf Artif Intell Statist – year: 2016 ident: ref102 article-title: Wild variational approximations publication-title: Proc Conf Neural Inf Process Syst Workshops – ident: ref79 doi: 10.1023/A:1007665907178 – start-page: 234 year: 2018 ident: ref8 article-title: Efficient gradient-free variational inference using policy search publication-title: Proc 35th Int Conf Mach Learn – start-page: 1218 year: 2015 ident: ref169 article-title: Markov Chain Monte Carlo and variational inference: Bridging the gap publication-title: Proceedings of the 32nd Intl Conf on Machine Learning – year: 2000 ident: ref193 article-title: The information bottleneck method publication-title: Arxiv Preprint Physics/0004057 – ident: ref172 doi: 10.1613/jair.251 – year: 2018 ident: ref139 article-title: Debiasing evidence approximations: On importance-weighted autoencoders and jackknife variational inference publication-title: Proc Int Conf Learn Representations – start-page: 2378 year: 2016 ident: ref109 article-title: Stein variational gradient descent: A general purpose Bayesian inference algorithm publication-title: Proc 30th Int Conf Neural Inf Process Syst – year: 2014 ident: ref126 article-title: Infer.NET 2.6 – start-page: 225 year: 1996 ident: ref144 article-title: A mean field algorithm for Bayes learning in large feed-forward neural networks publication-title: Proc 9th Int Conf Neural Inf Process – start-page: 647 year: 2016 ident: ref167 article-title: Overdispersed black-box variational inference publication-title: Proc Int Conf Uncertainty Artificial Intell – start-page: 489 year: 2017 ident: ref132 article-title: Reparameterization gradients through acceptance-rejection sampling algorithms publication-title: Proc Int Conf Artif Intell Statist – start-page: 3402 year: 2015 ident: ref82 article-title: Kullback-Leibler proximal variational inference publication-title: Proc 28th Int Conf Neural Inf Process Syst – year: 1988 ident: ref147 publication-title: Statistical Field Theory – start-page: 1070 year: 2016 ident: ref43 article-title: Early stopping as nonparametric variational inference publication-title: Proc 19th Int Conf Artif Intell – ident: ref5 doi: 10.1162/089976698300017746 – year: 2017 ident: ref111 article-title: The concrete distribution: A continuous relaxation of discrete random variables publication-title: ICLRE – ident: ref93 doi: 10.1214/aoms/1177729694 – year: 2017 ident: ref31 article-title: Reinterpreting importance-weighted autoencoders publication-title: Proc Workshop Int Conf Learn Represent – year: 2016 ident: ref228 article-title: Online spike-and-slab inference with stochastic expectation propagation publication-title: Proc 30th Conf Neural Inf Process Syst Workshops – ident: ref15 doi: 10.1080/01621459.2017.1285773 – ident: ref170 doi: 10.1214/13-BA858 – ident: ref18 doi: 10.1007/978-3-7908-2604-3_16 – ident: ref171 doi: 10.1162/089976601750265045 – start-page: 2323 year: 2015 ident: ref101 article-title: Stochastic expectation propagation publication-title: Proc 28th Int Conf Neural Inf Process Syst – start-page: 467 year: 1999 ident: ref131 article-title: Loopy belief propagation for approximate inference: An empirical study publication-title: Proc 15th Int Conf Uncertainty Artif Intell – year: 2017 ident: ref57 article-title: PixelVAE: A latent variable model for natural images publication-title: ICLRE – year: 2016 ident: ref220 article-title: Structured representation using latent variable models – start-page: 1462 year: 2015 ident: ref56 article-title: Draw: A recurrent neural network for image generation publication-title: Proceedings of the 32nd Intl Conf on Machine Learning – ident: ref17 doi: 10.1145/1143844.1143859 – ident: ref151 doi: 10.1145/1401890.1401960 – start-page: 361 year: 2015 ident: ref66 article-title: Structured stochastic variational inference publication-title: Proc Int Conf Artif Intell Statist – year: 2017 ident: ref75 article-title: Categorical reparameterization with gumbel-softmax publication-title: Proc Int Conf Learn Representations – year: 2014 ident: ref86 article-title: Stochastic gradient VB and the variational auto-encoder publication-title: Proc Int Conf Learn Representations – ident: ref204 doi: 10.1109/TIT.2005.850091 – start-page: 2438 year: 2014 ident: ref112 article-title: Smoothed Gradients for stochastic variational inference publication-title: Proc 27th Int Conf Neural Inf Process Syst – year: 2015 ident: ref91 article-title: Deep Kalman filters publication-title: Proc Conf Neural Inf Process Syst Workshops – year: 2017 ident: ref176 article-title: ZhuSuan: A library for Bayesian deep learning publication-title: arXiv 1709 05870 – start-page: 829 year: 2016 ident: ref60 article-title: Variational Gaussian copula inference publication-title: Proc 19th Int Conf Artif Intell – start-page: 1889 year: 2015 ident: ref173 article-title: Trust region policy optimization publication-title: Proc 31st Int Conf Mach Learn – ident: ref38 doi: 10.1109/CVPR.2017.637 – start-page: 1727 year: 2016 ident: ref122 article-title: Neural variational inference for text processing publication-title: Proc 33rd Int Conf Mach Learn – year: 2017 ident: ref2 article-title: Deep variational information bottleneck publication-title: Proc Int Conf Learn Representations – volume: 76 start-page: 1 year: 2016 ident: ref28 article-title: Stan: A probabilistic programming language publication-title: J Statist Software – year: 1985 ident: ref4 publication-title: Differential-Geometrical Methods in Statistics doi: 10.1007/978-1-4612-5056-2 – start-page: 3438 year: 2015 ident: ref90 article-title: Bayesian dark knowledge publication-title: Adv Neural Info Process Syst – year: 2016 ident: ref129 article-title: Learning in implicit generative models publication-title: arXiv 1610 03483 – volume: 14 start-page: 1303 year: 2013 ident: ref67 article-title: Stochastic variational inference publication-title: Journal of Machine Learning Research – year: 2017 ident: ref180 article-title: Autoencoding variational inference for topic models publication-title: Proc Int Conf Learn Representations – year: 2017 ident: ref12 article-title: Structured black box variational inference for latent time series models publication-title: Proc Int'l Conf Machine Learning Workshop – start-page: 704 year: 2016 ident: ref115 article-title: Variational Tempering publication-title: Proc 19th Int Conf Artif Intell – start-page: 1530 year: 2015 ident: ref159 article-title: Variational inference with normalizing flows publication-title: Proceedings of the 32nd Intl Conf on Machine Learning – year: 1995 ident: ref229 article-title: Information geometric measurements of generalisation – start-page: 567 year: 2009 ident: ref195 article-title: Variational learning of inducing variables in sparse gaussian processes publication-title: Proc 12th Int Conf Artif Intell Statist – year: 2016 ident: ref58 article-title: Boosting variational inference publication-title: NeurIPS WS Adv Approx Bayesian Inference – ident: ref97 doi: 10.1016/j.patcog.2011.10.004 – ident: ref110 doi: 10.1201/9781315366951 – ident: ref190 doi: 10.1080/14786437708235992 – start-page: 380 year: 2017 ident: ref11 article-title: Dynamic word embeddings publication-title: Proc 34th Int Conf Mach Learn – year: 2004 ident: ref125 article-title: Power EP – start-page: 1257 year: 2006 ident: ref177 article-title: Sparse Gaussian processes using pseudo-inputs publication-title: Proc 18th Int Conf Neural Inf Process Syst – volume: 19 start-page: 1 year: 2018 ident: ref32 article-title: Importance sampling for minibatches publication-title: JMLR – start-page: 856 year: 2010 ident: ref65 article-title: Online learning for Latent Dirichlet Allocation publication-title: Proc 23rd Int Conf Neural Inf Process Syst – year: 2015 ident: ref87 article-title: Adam: A method for stochastic optimization publication-title: Proc Int Conf Learn Representations – start-page: 1363 year: 2012 ident: ref146 article-title: Variational Bayesian inference with stochastic search publication-title: Proc 29th Int Conf Mach Learn – year: 2016 ident: ref53 publication-title: Deep Learning – ident: ref20 doi: 10.1016/0304-405X(77)90005-8 – year: 2015 ident: ref141 article-title: Perturbation theory for variational inference publication-title: Proc Conf Neural Inf Process Syst Workshops – ident: ref184 doi: 10.1007/s11222-015-9618-x – ident: ref215 doi: 10.1023/A:1022672621406 – year: 2015 ident: ref138 article-title: Embarrassingly parallel variational inference in nonconjugate models publication-title: arXiv 1510 04163 – volume: 11 start-page: 3235 year: 2010 ident: ref69 article-title: Approximate Riemannian conjugate gradient learning for fixed-form variational bayes publication-title: Journal of Machine Learning Research – volume: 14 start-page: 2857 year: 2013 ident: ref142 article-title: Perturbative corrections for approximate inference in gaussian latent variable models publication-title: Journal of Machine Learning Research – year: 0 ident: ref121 article-title: Fast and scalable Bayesian deep learning by weight-perturbation in adam publication-title: arXiv 1806 04854 – ident: ref7 doi: 10.1561/2200000052 – ident: ref96 doi: 10.1214/ss/1177013621 – year: 2018 ident: ref106 article-title: Variational message passing with structured inference networks publication-title: arXiv 1803 05589 – year: 1993 ident: ref136 article-title: Probabilistic inference using Markov Chain Monte Carlo methods – ident: ref44 doi: 10.1016/S0959-440X(96)80056-X – start-page: 201 year: 2003 ident: ref13 article-title: The IM algorithm: A variational approach to information maximization publication-title: Proc 16th Int Conf Neural Inf Process Syst – start-page: 1594 year: 2015 ident: ref175 article-title: An empirical study of stochastic variational algorithms for the beta bernoulli process publication-title: Proceedings of the 32nd Intl Conf on Machine Learning – start-page: 2249 year: 2012 ident: ref55 article-title: Scalable inference of overlapping communities publication-title: Proc 25th Int Conf Neural Inf Process Syst – volume: 36 start-page: 517 year: 2014 ident: ref50 article-title: Amortized inference in probabilistic reasoning publication-title: Proc 3rd Annu Meeting of the Cogn Sci Soc – start-page: 270 year: 2006 ident: ref83 article-title: Fast variational inference for Gaussian process models through KL-correction publication-title: Proc 17th Eur Conf Mach Learn – start-page: 351 year: 1999 ident: ref186 article-title: A theory of mean field approximation publication-title: Proc Int Conf Neural Inf Process – year: 2017 ident: ref41 article-title: Density estimation using real NVP publication-title: Proc Int Conf Learn Representations – start-page: 3564 year: 2015 ident: ref199 article-title: Copula variational inference publication-title: Proc 28th Int Conf Neural Inf Process Syst – start-page: 207 year: 2013 ident: ref99 article-title: Variational policy search via trajectory optimization publication-title: Proc 26th Int Conf Neural Inf Process Syst – year: 2012 ident: ref218 article-title: ADADELTA: An adaptive learning rate method publication-title: arXiv 1212 5701 – year: 2016 ident: ref200 article-title: Edward: A library for probabilistic modeling, inference, and criticism publication-title: arXiv 1610 09787 – ident: ref219 doi: 10.1145/2187836.2187955 – year: 2014 ident: ref85 article-title: Auto-encoding variational Bayes publication-title: Proc Int Conf Learn Representations – start-page: 3684 year: 2017 ident: ref211 article-title: Variational policy for guiding point processes publication-title: Proc 34th Int Conf Mach Learn – year: 2016 ident: ref210 article-title: Learning to draw samples: With application to amortized MLE for generative adversarial learning publication-title: arXiv 1611 01722 – start-page: 1353 year: 2006 ident: ref188 article-title: A collapsed variational Bayesian inference algorithm for latent Dirichlet allocation publication-title: Proc 19th Int Conf Neural Inf Process Syst – start-page: 272 year: 2013 ident: ref63 article-title: Gaussian processes for big data publication-title: Proc Int Conf Uncertainty Artif Intell – year: 0 ident: ref27 publication-title: Asymptotic Methods in Statistical Decision Theory – ident: ref3 doi: 10.1111/j.2517-6161.1966.tb00626.x – start-page: 1 year: 2018 ident: ref213 article-title: Frequentist consistency of variational bayes publication-title: JASA – start-page: 2420 year: 2017 ident: ref124 article-title: Variational boosting: Iteratively refining posterior approximations publication-title: Proc 34th Int Conf Mach Learn – year: 2017 ident: ref134 article-title: Stick-breaking variational autoencoders publication-title: Proc Int Conf Learn Representations – start-page: 803 year: 2003 ident: ref71 article-title: Online variational Bayesian learning publication-title: ICA – start-page: 6925 year: 2017 ident: ref162 article-title: Sticking the landing: An asymptotically zero-variance gradient estimator for variational inference publication-title: Proc Int Conf Neural Inf Process – year: 2016 ident: ref34 article-title: Variational auto-encoded deep Gaussian processes publication-title: Proc Int Conf Learn Representations – start-page: 207 year: 2013 ident: ref35 article-title: Deep Gaussian processes publication-title: Proc 16th Int Conf Artif Intell Statist – start-page: 1701 year: 2011 ident: ref88 article-title: Non-conjugate variational message passing for multinomial and binary regression publication-title: Proc 24th Int Conf Neural Inf Process Syst – start-page: 496 year: 2016 ident: ref155 article-title: Operator variational inference publication-title: Proc 30th Int Conf Neural Inf Process Syst – year: 2011 ident: ref62 article-title: Approximate inference in graphical models – volume: 4 start-page: 26 year: 2012 ident: ref191 article-title: Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude publication-title: Neural Netw Mach Learning – year: 2006 ident: ref165 publication-title: Simulation – start-page: 554 year: 1998 ident: ref185 article-title: Estimation of third-order correlations within mean field approximation publication-title: Proc 5th Int Conf Neural Inf Process Syst – ident: ref192 doi: 10.2307/2287970 – volume: 1 year: 1998 ident: ref183 publication-title: Reinforcement Learning An Introduction – start-page: 1081 year: 2016 ident: ref103 article-title: Rényi divergence variational inference publication-title: Proc 30th Int Conf Neural Inf Process Syst – year: 2005 ident: ref174 article-title: Expectation propagation for exponential families – start-page: 668 year: 2018 ident: ref23 article-title: Quasi-Monte Carlo Variational Inference publication-title: Proc 35th Int Conf Mach Learn – start-page: 4277 year: 2018 ident: ref153 article-title: Tighter variational bounds are not necessarily better publication-title: Proc 35th Int Conf Mach Learn – start-page: 2627 year: 2017 ident: ref203 article-title: REBAR: Low-variance, unbiased gradient estimates for discrete latent variable models publication-title: Proc 31st Int Conf Neural Inf Process Syst – ident: ref187 doi: 10.1162/089976600300015213 – year: 2016 ident: ref25 article-title: Importance weighted autoencoders publication-title: ICLRE – start-page: 2946 year: 2016 ident: ref77 article-title: Structured VAEs: Composing probabilistic graphical models and variational autoencoders publication-title: Proc Int Conf Neural Inf Process – volume: 9 year: 1990 ident: ref120 article-title: Spin glass theory and beyond – ident: ref36 doi: 10.1162/neco.1995.7.5.889 – start-page: 3257 year: 2014 ident: ref49 article-title: Distributed variational inference in sparse gaussian process regression and latent variable models publication-title: Proc 27th Int Conf Neural Inf Process Syst – start-page: 444 year: 2015 ident: ref92 article-title: Population empirical Bayes publication-title: Proc Int Conf Uncertainty Artificial Intell – start-page: 1903 year: 2009 ident: ref206 article-title: Simultaneous image classification and annotation publication-title: Proc IEEE Conf Comput Vis Pattern Recognit – start-page: 2796 year: 2007 ident: ref94 article-title: Collapsed variational Dirichlet process mixture models publication-title: Proc 20th Int Joint Conf Artif Intell – year: 2017 ident: ref227 article-title: Towards deeper understanding of variational autoencoding models publication-title: CoRR – start-page: 869 year: 2017 ident: ref148 article-title: Faster coordinate descent via adaptive importance sampling publication-title: Proc Int Conf Artif Intell Statist – start-page: 181 year: 2013 ident: ref208 article-title: Variance reduction for stochastic gradient optimization publication-title: Proc 26th Int Conf Neural Inf Process Syst – start-page: 5660 year: 2018 ident: ref217 article-title: Semi-implicit variational inference publication-title: Proc 35th Int Conf Mach Learn – year: 2017 ident: ref29 article-title: Variational lossy autoencoder publication-title: Proc Int Conf Learn Representations – start-page: 276 year: 2016 ident: ref108 article-title: A kernelized stein discrepancy for goodness-of-fit tests publication-title: Proc 33rd Int Conf Mach Learn – start-page: 1078 year: 2018 ident: ref30 article-title: Inference suboptimality in variational autoencoders publication-title: Proc 29th Int Conf Mach Learn – start-page: 1826 year: 2017 ident: ref119 article-title: The numerics of GANs publication-title: Proc 31st Int Conf Neural Inf Process Syst – start-page: 2888 year: 2012 ident: ref64 article-title: Fast variational inference in the conjugate exponential family publication-title: Proc 25th Int Conf Neural Inf Process Syst – volume: 12 start-page: 2121 year: 2011 ident: ref42 article-title: Adaptive subgradient methods for online learning and stochastic optimization publication-title: Journal of Machine Learning Research – start-page: 315 year: 2013 ident: ref78 article-title: Accelerating stochastic gradient descent using predictive variance reduction publication-title: Proc 26th Int Conf Neural Inf Process Syst – start-page: 147 year: 2006 ident: ref16 article-title: Correlated topic models publication-title: Proc 18th Int Conf Neural Inf Process Syst – start-page: 3708 year: 2017 ident: ref123 article-title: Reducing reparameterization gradient variance publication-title: Proc 31st Int Conf Neural Inf Process Syst – volume: 1050 year: 2016 ident: ref201 article-title: The Variational Gaussian Process publication-title: Stat – ident: ref135 doi: 10.1109/ICDMW.2007.33 – volume: 18 start-page: 1 year: 2016 ident: ref24 article-title: A unifying framework for sparse gaussian process approximation using power expectation propagation publication-title: JMLR – start-page: 2391 year: 2017 ident: ref118 article-title: Adversarial variational bayes: Unifying variational autoencoders and generative adversarial networks publication-title: Proc 34th Int Conf Mach Learn – start-page: ii-1854 year: 2014 ident: ref76 article-title: Stochastic variational inference for Bayesian time series models publication-title: Proc 31st Int Conf Mach Learn – start-page: 3646 year: 2017 ident: ref212 article-title: Robust Probabilistic Modeling with Bayesian Data Reweighting publication-title: Proc 34th Int Conf Mach Learn – start-page: 1309 year: 0 ident: ref168 article-title: Multimodal prediction and personalization of photo edits with deep generative models publication-title: AISTATS – start-page: 2568 year: 2016 ident: ref156 article-title: Hierarchical variational models publication-title: Proc 33rd Int Conf Mach Learn – start-page: 2732 year: 2017 ident: ref39 article-title: Variational inference via chi-upper bound minimization publication-title: Proc Int Conf Neural Inf Process – ident: ref140 doi: 10.1111/rssb.12185 – year: 2014 ident: ref158 article-title: Expectation propagation publication-title: arXiv 1409 6179 – ident: ref164 doi: 10.1016/0167-8655(90)90010-Y – year: 2015 ident: ref89 article-title: Stochastic gradient variational Bayes for gamma approximating distributions publication-title: ArXiv – year: 0 ident: ref202 article-title: Hierarchical implicit models and likelihood-free variational inference publication-title: arXiv 1702 08896 – start-page: 3599 year: 2014 ident: ref45 article-title: Stochastic variational inference for hidden Markov models publication-title: Proc 27th Int Conf Neural Inf Process Syst – start-page: 681 year: 2011 ident: ref214 article-title: Bayesian learning via stochastic gradient langevin dynamics publication-title: Proc 28th Int Conf Mach Learn – start-page: 583 year: 1972 ident: ref181 article-title: A bound for the error in the normal approximation to the distribution of a sum of dependent random variables publication-title: Proc Berkeley Symp Math Statist Probability – year: 2005 ident: ref128 article-title: Divergence measures and message passing – year: 2017 ident: ref223 article-title: Determinantal point processes for mini-batch diversification publication-title: Proc Int Conf Uncertainty Artif Intell – start-page: 2638 year: 2015 ident: ref194 article-title: Local expectation gradients for black box variational inference publication-title: Proc 28th Int Conf Neural Inf Process Syst – start-page: 718 year: 2017 ident: ref37 article-title: Automated inference using adaptive batch sizes publication-title: Proc Int Conf Artif Intell Statist – ident: ref130 doi: 10.2307/1428011 – ident: ref74 doi: 10.1007/978-94-011-5014-9_6 – ident: ref133 doi: 10.1162/neco.2007.19.4.1112 – ident: ref222 doi: 10.1109/ICCVW.2013.41 – year: 2017 ident: ref59 article-title: Stein variational adaptive importance sampling publication-title: Proc Int Conf Uncertainty Artif Intell – ident: ref26 doi: 10.1007/s10107-012-0572-5 – year: 0 ident: ref196 article-title: Learning model reparametrizations: Implicit variational inference by fitting MCMC distributions publication-title: arXiv 1708 01529 – volume: 1 start-page: 995 year: 1987 ident: ref149 article-title: A mean field theory learning algorithm for neural networks publication-title: Complex Syst – volume: 18 start-page: 4873 year: 2017 ident: ref114 article-title: Stochastic gradient descent as approximate Bayesian inference publication-title: Journal of Machine Learning Research – year: 2009 ident: ref179 article-title: On integral probability metrics,$\backslash$?phi-divergences and binary classification publication-title: arXiv preprint arXiv 0901 2698 – start-page: 28 year: 1994 ident: ref9 article-title: Laplace's method approximations for probabilistic inferencein belief networks with continuous variables publication-title: Proc 10th Conf Uncertainty in Artif Intell – year: 2016 ident: ref178 article-title: How to train deep variational autoencoders and probabilistic ladder networks publication-title: Proc 33rd Int Conf Mach Learn – year: 2006 ident: ref14 publication-title: Pattern Recognition and Machine Learning – ident: ref150 doi: 10.1088/0305-4470/15/6/035 – start-page: 2672 year: 2014 ident: ref54 article-title: Generative adversarial nets publication-title: Proc Int Conf Neural Inf Process – ident: ref19 doi: 10.18653/v1/K16-1002 – start-page: 3403 year: 2018 ident: ref116 article-title: Iterative amortized inference publication-title: Proc Int Conf Mach Learn – volume: 6 start-page: 661 year: 2005 ident: ref216 article-title: Variational message passing publication-title: Journal of Machine Learning Research – year: 2018 ident: ref100 article-title: Approximate inference: New visions – year: 2017 ident: ref72 article-title: Variational inference using implicit distributions publication-title: arXiv 1702 08235 – year: 2008 ident: ref61 article-title: Parameter estimation for text analysis – volume: 14 start-page: 1005 year: 2013 ident: ref207 article-title: Variational inference in non-conjugate models publication-title: Journal of Machine Learning Research – volume: 8 start-page: 95 year: 1964 ident: ref33 article-title: Eine informationstheoretische ungleichung und ihre anwendung auf beweis der ergodizitaet von markoffschen ketten publication-title: Magyer Tud Akad Mat Kutato Int Koezl – year: 2017 ident: ref10 article-title: Coupling adaptive batch sizes with learning rates publication-title: Proc Int Conf Uncertainty Artif Intell – start-page: 416 year: 2011 ident: ref95 article-title: Approximate inference for the loss-calibrated Bayesian publication-title: Proc 14th Int Conf Artif Intell Statist – year: 2017 ident: ref163 article-title: Discrete variational autoencoders publication-title: Proc Int Conf Learn Representations – year: 2014 ident: ref225 article-title: Accelerating minibatch stochastic gradient descent using stratified sampling publication-title: arXiv 1405 3080 – ident: ref6 doi: 10.1109/TIT.2009.2030485 – year: 2016 ident: ref81 article-title: Adversarial message passing for graphical models publication-title: Proc 30th Conf Neural Inf Process Syst Workshops – start-page: 663 year: 2012 ident: ref51 article-title: Nonparametric variational inference publication-title: Proc 29th Int Conf Mach Learn – year: 2017 ident: ref105 article-title: Approximate inference with amortised MCMC publication-title: arXiv 1702 08343 – start-page: 1153 year: 2015 ident: ref117 article-title: The population posterior and Bayesian modeling on streams publication-title: Proc 28th Int Conf Neural Inf Process Syst – start-page: ii-1278 year: 2014 ident: ref160 article-title: Stochastic backpropagation and approximate inference in deep generative models publication-title: Proc 31st Int Conf Mach Learn – start-page: 460 year: 2016 ident: ref166 article-title: The generalized reparameterization gradient publication-title: Proc 30th Int Conf Neural Inf Process Syst – volume: 1 start-page: 1 year: 2008 ident: ref205 article-title: Graphical models, exponential families, and variational inference publication-title: Foundations & Trends in Mach Learn doi: 10.1561/2200000001 – year: 2015 ident: ref40 article-title: NICE: Non-linear independent components estimation publication-title: ICLR-WS – ident: ref46 doi: 10.1137/110830629 – start-page: 841 year: 2017 ident: ref47 article-title: CPSG-MCMC: Clustering-based preprocessing method for stochastic gradient MCMC publication-title: Proc Int Conf Artif Intell Statist – start-page: 841 year: 2011 ident: ref197 article-title: Variational heteroscedastic Gaussian process regression publication-title: Proc 28th Int Conf Mach Learn – year: 0 ident: ref98 article-title: Reinforcement learning and control as probabilistic inference: Tutorial and review publication-title: arXiv 1805 00909 – start-page: 220 year: 2001 ident: ref80 article-title: Second order approximations for probability models publication-title: Proc 13th Int Conf Neural Inf Process Syst – ident: ref143 doi: 10.7551/mitpress/1100.001.0001 – start-page: 4267 year: 2018 ident: ref152 article-title: On Nesting Monte Carlo Estimators publication-title: Proc 35th Int Conf Mach Learn – start-page: 305 year: 2007 ident: ref70 article-title: Natural conjugate gradient in variational inference publication-title: Proc Int Conf Neural Inf Process – ident: ref198 doi: 10.1145/3064899.3064910 – ident: ref182 doi: 10.1109/TPAMI.2008.157 – year: 1995 ident: ref68 publication-title: Introduction to Mathematical Statistics – start-page: 5079 year: 2017 ident: ref221 article-title: Perturbative black box variational inference publication-title: Proc 31st Int Conf Neural Inf Process Syst – year: 2016 ident: ref48 article-title: Uncertainty in deep learning – start-page: 4743 year: 2016 ident: ref84 article-title: Improving variational autoencoders with inverse autoregressive flow publication-title: Proc 30th Int Conf Neural Inf Process Syst – volume: 521 start-page: 452 year: 2015 ident: ref52 article-title: Probabilistic machine learning and artificial intelligence publication-title: Nature doi: 10.1038/nature14541 – start-page: 528 year: 1996 ident: ref73 article-title: Fast learning by bounding likelihoods in sigmoid type belief networks publication-title: Proc 8th Int Conf Neural Inf Process Syst – start-page: 2503 year: 2015 ident: ref189 article-title: A trust-region method for stochastic variational inference with applications to streaming data publication-title: Proceedings of the 32nd Intl Conf on Machine Learning – year: 0 ident: ref224 article-title: Active mini-batch sampling using repulsive point processes publication-title: arXiv 1804 02772 |
| SSID | ssj0014503 |
| Score | 2.7182212 |
| Snippet | Many modern unsupervised or semi-supervised machine learning algorithms rely on Bayesian probabilistic models. These models are usually intractable and thus... |
| SourceID | swepub proquest pubmed crossref ieee |
| SourceType | Open Access Repository Aggregation Database Index Database Enrichment Source Publisher |
| StartPage | 2008 |
| SubjectTerms | Algorithms approximate Bayesian inference Bayes methods Bayesian analysis Computational modeling Hidden Markov models Inference inference networks Machine learning Optimization Probabilistic logic Probabilistic models reparameterization gradients scalable inference Stochastic processes structured variational approximations Variational inference |
| Title | Advances in Variational Inference |
| URI | https://ieeexplore.ieee.org/document/8588399 https://www.ncbi.nlm.nih.gov/pubmed/30596568 https://www.proquest.com/docview/2250718828 https://www.proquest.com/docview/2162494415 https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-255405 |
| Volume | 41 |
| WOSCitedRecordID | wos000473598800016&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 2160-9292 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014503 issn: 0162-8828 databaseCode: RIE dateStart: 19790101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8QwEB5URPTg-ra-qCBetNom6SY5Lj5QUPGgy95Cm6S4KF1xd_39TtIHIiJ4K01a0nwznZkk8w3AUa4zg2ZJR0YQFqE95pHMSRbpwpiYWJMnnkynf8cfHsRgIB9n4LTNhbHW-sNn9sxd-r18M9JTt1R2LlKB9lzOwiznvMrVancMWOqrIKMHgxqOYUSTIBPL86fH3v2tO8UlzogQzuFZhAXq6s6kjmH1mz3yBVZ-8zV_EIl643Pd-d-wV2C5djLDXiUVqzBjyzXoNAUcwlqf12DpGxvhOhz2qvMA43BYhn2Moet1wvC2SQrcgOfrq6eLm6iuoBBpRukkSmRhKC94ZqXMWZHY1BLGC-pYuhxC1LHRUWsTnXW7CAyh3KBV04XImNaW0k2YK0el3YZQG11IxkjMMsFMWuTCGJLlkuDt2CQ2gKSZR6VrenFX5eJN-TAjlsrDoBwMqoYhgJP2mfeKXOPP3utubtue9bQGsNfApWr9Gyv8S6Hv5OQggMO2GTXHbYdkpR1NsQ-KCpMungxgq4K5fXcjHQEcV7i3LY6O-3LY7ymEWL1OXhTGZOj07vw-uF1YxE-Q1YnBPZibfEztPszrz8lw_HGAcjwQB16OvwAcQetv |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dT9swED8BQ3w8DAYbhLGRSogXCCS2Q-zHaBtqRan60FW8WYntaBVTimi6v39n50MVQpP2FsVO5Ph3l7uzz78DOM9VptEsqUBzwgK0x0kgcpIFqtA6JEbnkSPTmQ6T0Yg_PorxGlx1Z2GMMS75zFzbS7eXr-dqaZfKbnjM0Z6LdXgXM0ai-rRWt2fAYlcHGX0Y1HEMJNojMqG4mYzTh4HN4-LXhHPr8uzAFrWVZ2LLsbpikVyJlbe8zVdUos783O3938D34X3jZvppLRcfYM2UB7DXlnDwG40-gN0VPsJD6KV1RsDCn5X-FKPoZqXQH7THAj_Cz7sfk2_9oKmhEChGaRVEotA0KZLMCJGzIjKxISwpqOXpshhRy0dHjYlUdnuL0BCaaLRrquAZU8pQ-gk2ynlpjsFXWhUC5z1kGWc6LnKuNclyQfB2qCPjQdTOo1QNwbitc_FbukAjFNLBIC0MsoHBg8vumeeaXuOfvQ_t3HY9m2n14LSFSzYauJD4n0LvycqBB72uGXXHbohkpZkvsQ-KChM2ovTgqIa5e3crHR5c1Lh3LZaQ-_tsmkqEWD5VvyRGZej2nrw9uDPY7k8ehnI4GN1_hh38HFHnD57CRvWyNF9gU_2pZouXr06a_wL1lO3O |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Advances+in+Variational+Inference&rft.jtitle=IEEE+transactions+on+pattern+analysis+and+machine+intelligence&rft.au=Zhang%2C+Cheng&rft.au=Butepage%2C+Judith&rft.au=Kjellstrom%2C+Hedvig&rft.au=Mandt%2C+Stephan&rft.date=2019-08-01&rft.eissn=1939-3539&rft.volume=41&rft.issue=8&rft.spage=2008&rft_id=info:doi/10.1109%2FTPAMI.2018.2889774&rft_id=info%3Apmid%2F30596568&rft.externalDocID=30596568 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0162-8828&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0162-8828&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0162-8828&client=summon |