PrediRep: Modeling hierarchical predictive coding with an unsupervised deep learning network

Hierarchical predictive coding (hPC) provides a compelling framework for understanding how the cortex predicts future sensory inputs by minimizing prediction errors through an internal generative model of the external world. Existing deep learning models inspired by hPC incorporate architectural cho...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Neural networks Jg. 185; S. 107246
Hauptverfasser: Hashim, Ibrahim C., Senden, Mario, Goebel, Rainer
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States Elsevier Ltd 01.05.2025
Schlagworte:
ISSN:0893-6080, 1879-2782, 1879-2782
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Hierarchical predictive coding (hPC) provides a compelling framework for understanding how the cortex predicts future sensory inputs by minimizing prediction errors through an internal generative model of the external world. Existing deep learning models inspired by hPC incorporate architectural choices that deviate from core hPC principles, potentially limiting their utility for neuroscientific investigations. We introduce PrediRep (Predicting Representations), a novel deep learning network that adheres more closely to architectural principles of hPC. We validate PrediRep by comparing its functional alignment with hPC to that of existing models after being trained on a next-frame prediction task. Our findings demonstrate that PrediRep, particularly when trained with an all-level loss function (PrediRepAll), exhibits high functional alignment with hPC. In contrast to other contemporary deep learning networks inspired by hPC, it consistently processes input-relevant information at higher hierarchical levels and maintains active representations and accurate predictions across all hierarchical levels. Although PrediRep was designed primarily to serve as a model suitable for neuroscientific research rather than to optimize performance, it nevertheless achieves competitive performance in next-frame prediction while utilizing significantly fewer trainable parameters than alternative models. Our results underscore that even minor architectural deviations from neuroscientific theories like hPC can lead to significant functional discrepancies. By faithfully adhering to hPC principles, PrediRep provides a more accurate tool for in silico exploration of cortical phenomena. PrediRep’s lightweight and biologically plausible design makes it well-suited for future studies aiming to investigate the neural underpinnings of predictive coding and to derive empirically testable predictions.
AbstractList Hierarchical predictive coding (hPC) provides a compelling framework for understanding how the cortex predicts future sensory inputs by minimizing prediction errors through an internal generative model of the external world. Existing deep learning models inspired by hPC incorporate architectural choices that deviate from core hPC principles, potentially limiting their utility for neuroscientific investigations. We introduce PrediRep (Predicting Representations), a novel deep learning network that adheres more closely to architectural principles of hPC. We validate PrediRep by comparing its functional alignment with hPC to that of existing models after being trained on a next-frame prediction task. Our findings demonstrate that PrediRep, particularly when trained with an all-level loss function (PrediRepAll), exhibits high functional alignment with hPC. In contrast to other contemporary deep learning networks inspired by hPC, it consistently processes input-relevant information at higher hierarchical levels and maintains active representations and accurate predictions across all hierarchical levels. Although PrediRep was designed primarily to serve as a model suitable for neuroscientific research rather than to optimize performance, it nevertheless achieves competitive performance in next-frame prediction while utilizing significantly fewer trainable parameters than alternative models. Our results underscore that even minor architectural deviations from neuroscientific theories like hPC can lead to significant functional discrepancies. By faithfully adhering to hPC principles, PrediRep provides a more accurate tool for in silico exploration of cortical phenomena. PrediRep's lightweight and biologically plausible design makes it well-suited for future studies aiming to investigate the neural underpinnings of predictive coding and to derive empirically testable predictions.Hierarchical predictive coding (hPC) provides a compelling framework for understanding how the cortex predicts future sensory inputs by minimizing prediction errors through an internal generative model of the external world. Existing deep learning models inspired by hPC incorporate architectural choices that deviate from core hPC principles, potentially limiting their utility for neuroscientific investigations. We introduce PrediRep (Predicting Representations), a novel deep learning network that adheres more closely to architectural principles of hPC. We validate PrediRep by comparing its functional alignment with hPC to that of existing models after being trained on a next-frame prediction task. Our findings demonstrate that PrediRep, particularly when trained with an all-level loss function (PrediRepAll), exhibits high functional alignment with hPC. In contrast to other contemporary deep learning networks inspired by hPC, it consistently processes input-relevant information at higher hierarchical levels and maintains active representations and accurate predictions across all hierarchical levels. Although PrediRep was designed primarily to serve as a model suitable for neuroscientific research rather than to optimize performance, it nevertheless achieves competitive performance in next-frame prediction while utilizing significantly fewer trainable parameters than alternative models. Our results underscore that even minor architectural deviations from neuroscientific theories like hPC can lead to significant functional discrepancies. By faithfully adhering to hPC principles, PrediRep provides a more accurate tool for in silico exploration of cortical phenomena. PrediRep's lightweight and biologically plausible design makes it well-suited for future studies aiming to investigate the neural underpinnings of predictive coding and to derive empirically testable predictions.
Hierarchical predictive coding (hPC) provides a compelling framework for understanding how the cortex predicts future sensory inputs by minimizing prediction errors through an internal generative model of the external world. Existing deep learning models inspired by hPC incorporate architectural choices that deviate from core hPC principles, potentially limiting their utility for neuroscientific investigations. We introduce PrediRep (Predicting Representations), a novel deep learning network that adheres more closely to architectural principles of hPC. We validate PrediRep by comparing its functional alignment with hPC to that of existing models after being trained on a next-frame prediction task. Our findings demonstrate that PrediRep, particularly when trained with an all-level loss function (PrediRepAll), exhibits high functional alignment with hPC. In contrast to other contemporary deep learning networks inspired by hPC, it consistently processes input-relevant information at higher hierarchical levels and maintains active representations and accurate predictions across all hierarchical levels. Although PrediRep was designed primarily to serve as a model suitable for neuroscientific research rather than to optimize performance, it nevertheless achieves competitive performance in next-frame prediction while utilizing significantly fewer trainable parameters than alternative models. Our results underscore that even minor architectural deviations from neuroscientific theories like hPC can lead to significant functional discrepancies. By faithfully adhering to hPC principles, PrediRep provides a more accurate tool for in silico exploration of cortical phenomena. PrediRep’s lightweight and biologically plausible design makes it well-suited for future studies aiming to investigate the neural underpinnings of predictive coding and to derive empirically testable predictions.
ArticleNumber 107246
Author Senden, Mario
Hashim, Ibrahim C.
Goebel, Rainer
Author_xml – sequence: 1
  givenname: Ibrahim C.
  orcidid: 0009-0008-4487-9180
  surname: Hashim
  fullname: Hashim, Ibrahim C.
  email: ibrahimhashim.coding@gmail.com
– sequence: 2
  givenname: Mario
  orcidid: 0000-0002-5598-6167
  surname: Senden
  fullname: Senden, Mario
  email: mario.senden@maastrichtuniversity.nl
– sequence: 3
  givenname: Rainer
  orcidid: 0000-0003-1780-2467
  surname: Goebel
  fullname: Goebel, Rainer
BackLink https://www.ncbi.nlm.nih.gov/pubmed/39946763$$D View this record in MEDLINE/PubMed
BookMark eNp9kE1r3DAQhkXYkGw-_kEpPvbirWTZlp1DoSz9gg0JpbkFhDwadbX1yq5k75J_HxknOeY0MPPMMO9zQRauc0jIB0ZXjLLy827lcHQ4rDKaFbElsrw8IUtWiTrNRJUtyJJWNU9LWtFzchHCjlJaVjk_I-e8rvNSlHxJHu89avsb-5vkttPYWvc32Vr0ysPWgmqTfprDYA-YQKen8dEO20S5ZHRh7NEfbECdaMQ-aVF5NyHxrWPn_12RU6PagNcv9ZI8fP_2Z_0z3dz9-LX-ukmBF2JIBVKjtKprwTkDAyxrjBFQNGVTMEBagGlyajKuBIUMAETORQFYaMqFUTm_JJ_mu73v_o8YBrm3AbBtlcNuDJKzMqYtaF5H9OMLOjZ71LL3dq_8k3w1EoF8BsB3IXg0bwijchIvd3IWLyfxchYf177MaxhzHqJAGcCigyjPIwxSd_b9A8_qeo8o
Cites_doi 10.3389/fpsyg.2016.01792
10.1364/JOSAA.20.001434
10.3389/fpsyg.2018.00345
10.1016/j.visres.2023.108195
10.1038/s41583-020-0277-3
10.1093/cercor/1.1.1
10.1016/j.cub.2015.12.038
10.1017/S0140525X12000477
10.1016/j.bandc.2015.11.003
10.1016/j.visres.2008.03.009
10.1038/s42256-020-0170-9
10.1002/cne.23458
10.1007/BF00198477
10.1113/jphysiol.1968.sp008455
10.1016/j.cub.2015.08.057
10.1177/0278364913491297
10.1016/j.neuroimage.2020.117479
10.1002/wcs.142
10.1016/j.neuron.2012.10.038
10.1038/4580
10.1038/ncomms13276
10.1016/j.tins.2022.09.007
10.1371/journal.pcbi.1010719
10.1109/TIP.2003.819861
10.1016/j.schres.2020.10.009
10.3389/fncom.2021.666131
10.1109/5.58337
10.1523/JNEUROSCI.5003-11.2012
10.1016/j.neunet.2003.06.005
10.1523/JNEUROSCI.0852-19.2019
10.1371/journal.pbio.3001023
10.1162/NECO_a_00222
ContentType Journal Article
Copyright 2025 The Authors
Copyright © 2025 The Authors. Published by Elsevier Ltd.. All rights reserved.
Copyright_xml – notice: 2025 The Authors
– notice: Copyright © 2025 The Authors. Published by Elsevier Ltd.. All rights reserved.
DBID 6I.
AAFTH
AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1016/j.neunet.2025.107246
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic

MEDLINE
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: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 1879-2782
ExternalDocumentID 39946763
10_1016_j_neunet_2025_107246
S089360802500125X
Genre Journal Article
GroupedDBID ---
--K
--M
-~X
.DC
.~1
0R~
123
186
1B1
1RT
1~.
1~5
29N
4.4
457
4G.
53G
5RE
5VS
6I.
6TJ
7-5
71M
8P~
9JM
9JN
AABNK
AAEDT
AAEDW
AAFTH
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AATTM
AAXKI
AAXLA
AAXUO
AAYFN
AAYWO
ABAOU
ABBOA
ABCQJ
ABDPE
ABEFU
ABFNM
ABFRF
ABHFT
ABIVO
ABJNI
ABLJU
ABMAC
ABWVN
ABXDB
ACDAQ
ACGFO
ACGFS
ACIUM
ACNNM
ACRLP
ACRPL
ACVFH
ACZNC
ADBBV
ADCNI
ADEZE
ADGUI
ADJOM
ADMUD
ADNMO
ADRHT
AEBSH
AECPX
AEFWE
AEIPS
AEKER
AENEX
AEUPX
AFJKZ
AFPUW
AFTJW
AFXIZ
AGCQF
AGHFR
AGQPQ
AGRNS
AGUBO
AGWIK
AGYEJ
AHHHB
AHJVU
AHZHX
AIALX
AIEXJ
AIGII
AIIUN
AIKHN
AITUG
AKBMS
AKRWK
AKYEP
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
AOUOD
APXCP
ARUGR
ASPBG
AVWKF
AXJTR
AZFZN
BJAXD
BKOJK
BLXMC
BNPGV
CS3
DU5
EBS
EFJIC
EJD
EO8
EO9
EP2
EP3
F0J
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-2
G-Q
GBLVA
GBOLZ
HLZ
HMQ
HVGLF
HZ~
IHE
J1W
JJJVA
K-O
KOM
KZ1
LG9
LMP
M2V
M41
MHUIS
MO0
MOBAO
MVM
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
R2-
RIG
ROL
RPZ
SBC
SCC
SDF
SDG
SDP
SES
SEW
SNS
SPC
SPCBC
SSH
SSN
SST
SSV
SSW
SSZ
T5K
TAE
UAP
UNMZH
VOH
WUQ
XPP
ZMT
~G-
9DU
AAYXX
ACLOT
CITATION
EFKBS
EFLBG
~HD
AACTN
CGR
CUY
CVF
ECM
EIF
NPM
7X8
ID FETCH-LOGICAL-c357t-7e0fada997331cfc12bff7c5b6b51ce05cfb40f23a70c2ccc74375ce5d037fa43
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001427277300001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0893-6080
1879-2782
IngestDate Sun Sep 28 05:14:32 EDT 2025
Fri May 09 01:30:37 EDT 2025
Sat Nov 29 06:52:28 EST 2025
Sat Jun 07 17:02:19 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords Deep learning
Predictive coding
Predictive processing
Temporal prediction
Unsupervised learning
Language English
License This is an open access article under the CC BY license.
Copyright © 2025 The Authors. Published by Elsevier Ltd.. All rights reserved.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c357t-7e0fada997331cfc12bff7c5b6b51ce05cfb40f23a70c2ccc74375ce5d037fa43
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0002-5598-6167
0000-0003-1780-2467
0009-0008-4487-9180
OpenAccessLink https://dx.doi.org/10.1016/j.neunet.2025.107246
PMID 39946763
PQID 3166765049
PQPubID 23479
ParticipantIDs proquest_miscellaneous_3166765049
pubmed_primary_39946763
crossref_primary_10_1016_j_neunet_2025_107246
elsevier_sciencedirect_doi_10_1016_j_neunet_2025_107246
PublicationCentury 2000
PublicationDate May 2025
2025-05-00
2025-May
20250501
PublicationDateYYYYMMDD 2025-05-01
PublicationDate_xml – month: 05
  year: 2025
  text: May 2025
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle Neural networks
PublicationTitleAlternate Neural Netw
PublicationYear 2025
Publisher Elsevier Ltd
Publisher_xml – name: Elsevier Ltd
References Hubel, Wiesel (b16) 1968; 195
Dollar, Wojek, Schiele, Perona (b4) 2009
Hosseini, Maida (b14) 2020
Spratling (b42) 2012; 24
Millidge, Tang, Osanlouy, Bogacz (b29) 2023
Goroshin, Mathieu, LeCun (b11) 2015; Vol. 28
Geiger, Lenz, Stiller, Urtasun (b10) 2013; 32
Friston (b8) 2003; 16
Lillicrap, Santoro, Marris, Akerman, Hinton (b23) 2020; 21
Kirubeswaran, Storrs (b19) 2023; 206
O’Reilly, Wyatte, Rohrlich (b33) 2014
Dora, Bohte, Pennartz (b5) 2021; 15
Grauman, Westbury, Byrne, Chavis, Furnari, Girdhar, Hamburger, Jiang, Liu, Liu, Martin, Nagarajan, Radosavovic, Ramakrishnan, Ryan, Sharma, Wray, Xu, Xu, Malik (b12) 2022
Shi, Chen, Wang, Yeung, Wong, Woo (b38) 2015
Tang, Salvatori, Millidge, Song, Lukasiewicz, Bogacz (b46) 2023; 19
Vondrick, Pirsiavash, Torralba (b48) 2016; Vol. 29
Softky (b40) 1995; Vol. 8
Spratling (b41) 2008; 48
Patraucean, Handa, Cipolla (b35) 2015
Vezoli, Magrou, Goebel, Wang, Knoblauch, Vinck, Kennedy (b47) 2021; 225
Jiang, Rao (b17) 2022
Werbos (b52) 1990; 78
Aitken, Menelaou, Warrington, Koolschijn, Corbin, Callaghan, Kok (b1) 2020; 18
Friston (b9) 2005; 360
Markov, Vezoli, Chameau, Falchier, Quilodran, Huissoud, Lamy, Misery, Giroud, Ullman, Barone, Dehay, Knoblauch, Kennedy (b26) 2014; 522
Spratling (b43) 2017; 112
Lillicrap, Cownden, Tweed, Akerman (b22) 2016; 7
Morgan, Petro, Muckli (b30) 2019; 39
Rao, Ballard (b36) 1998; 9
Srivastava, Mansimov, Salakhutdinov (b44) 2015
Straka, Svoboda, Hoffmann (b45) 2023
Clark (b3) 2013; 36
Finn, Goodfellow, Levine (b7) 2016
Shipp (b39) 2016; 7
Mumford (b32) 1992; 66
Lee, Mumford (b21) 2003; 20
Mathieu, Couprie, LeCun (b27) 2015
Wang, Bovik, Sheikh, Simoncelli (b50) 2004; 13
Watanabe, Kitaoka, Sakamoto, Yasugi, Tanaka (b51) 2018; 9
Kingma, Ba (b18) 2014
Lotter, Kreiman, Cox (b24) 2016
Felleman, Van Essen (b6) 1991; 1
Palm (b34) 2012
Muckli, De Martino, Vizioli, Petro, Smith, Ugurbil, Goebel, Yacoub (b31) 2015; 25
Rao, Ballard (b37) 1999; 2
Lotter, Kreiman, Cox (b25) 2020; 2
Haarsma, Kok, Browning (b13) 2022; 245
Wacongne, Changeux, Dehaene (b49) 2012; 32
Bastos, Usrey, Adams, Mangun, Fries, Friston (b2) 2012; 76
Mikulasch, Rudelt, Wibral, Priesemann (b28) 2023; 46
Huang, Rao (b15) 2011; 2
Kok, Bains, van Mourik, Norris, de Lange (b20) 2016; 26
Lillicrap (10.1016/j.neunet.2025.107246_b23) 2020; 21
Vondrick (10.1016/j.neunet.2025.107246_b48) 2016; Vol. 29
Spratling (10.1016/j.neunet.2025.107246_b41) 2008; 48
Srivastava (10.1016/j.neunet.2025.107246_b44) 2015
Finn (10.1016/j.neunet.2025.107246_b7) 2016
Clark (10.1016/j.neunet.2025.107246_b3) 2013; 36
Haarsma (10.1016/j.neunet.2025.107246_b13) 2022; 245
Mikulasch (10.1016/j.neunet.2025.107246_b28) 2023; 46
Huang (10.1016/j.neunet.2025.107246_b15) 2011; 2
Friston (10.1016/j.neunet.2025.107246_b8) 2003; 16
Millidge (10.1016/j.neunet.2025.107246_b29) 2023
Hubel (10.1016/j.neunet.2025.107246_b16) 1968; 195
Lillicrap (10.1016/j.neunet.2025.107246_b22) 2016; 7
Lotter (10.1016/j.neunet.2025.107246_b25) 2020; 2
Wacongne (10.1016/j.neunet.2025.107246_b49) 2012; 32
Jiang (10.1016/j.neunet.2025.107246_b17) 2022
O’Reilly (10.1016/j.neunet.2025.107246_b33) 2014
Mathieu (10.1016/j.neunet.2025.107246_b27) 2015
Muckli (10.1016/j.neunet.2025.107246_b31) 2015; 25
Mumford (10.1016/j.neunet.2025.107246_b32) 1992; 66
Lotter (10.1016/j.neunet.2025.107246_b24) 2016
Geiger (10.1016/j.neunet.2025.107246_b10) 2013; 32
Wang (10.1016/j.neunet.2025.107246_b50) 2004; 13
Shi (10.1016/j.neunet.2025.107246_b38) 2015
Felleman (10.1016/j.neunet.2025.107246_b6) 1991; 1
Vezoli (10.1016/j.neunet.2025.107246_b47) 2021; 225
Softky (10.1016/j.neunet.2025.107246_b40) 1995; Vol. 8
Rao (10.1016/j.neunet.2025.107246_b37) 1999; 2
Markov (10.1016/j.neunet.2025.107246_b26) 2014; 522
Kirubeswaran (10.1016/j.neunet.2025.107246_b19) 2023; 206
Werbos (10.1016/j.neunet.2025.107246_b52) 1990; 78
Grauman (10.1016/j.neunet.2025.107246_b12) 2022
Tang (10.1016/j.neunet.2025.107246_b46) 2023; 19
Kingma (10.1016/j.neunet.2025.107246_b18) 2014
Spratling (10.1016/j.neunet.2025.107246_b42) 2012; 24
Watanabe (10.1016/j.neunet.2025.107246_b51) 2018; 9
Aitken (10.1016/j.neunet.2025.107246_b1) 2020; 18
Bastos (10.1016/j.neunet.2025.107246_b2) 2012; 76
Dollar (10.1016/j.neunet.2025.107246_b4) 2009
Dora (10.1016/j.neunet.2025.107246_b5) 2021; 15
Lee (10.1016/j.neunet.2025.107246_b21) 2003; 20
Palm (10.1016/j.neunet.2025.107246_b34) 2012
Kok (10.1016/j.neunet.2025.107246_b20) 2016; 26
Morgan (10.1016/j.neunet.2025.107246_b30) 2019; 39
Shipp (10.1016/j.neunet.2025.107246_b39) 2016; 7
Patraucean (10.1016/j.neunet.2025.107246_b35) 2015
Goroshin (10.1016/j.neunet.2025.107246_b11) 2015; Vol. 28
Hosseini (10.1016/j.neunet.2025.107246_b14) 2020
Friston (10.1016/j.neunet.2025.107246_b9) 2005; 360
Rao (10.1016/j.neunet.2025.107246_b36) 1998; 9
Straka (10.1016/j.neunet.2025.107246_b45) 2023
Spratling (10.1016/j.neunet.2025.107246_b43) 2017; 112
References_xml – volume: 20
  start-page: 1434
  year: 2003
  ident: b21
  article-title: Hierarchical Bayesian inference in the visual cortex
  publication-title: Journal of the Optical Society of America A
– year: 2023
  ident: b29
  article-title: Predictive coding networks for temporal prediction
  publication-title: Neuroscience
– volume: 7
  start-page: 13276
  year: 2016
  ident: b22
  article-title: Random synaptic feedback weights support error backpropagation for deep learning
  publication-title: Nature Communications
– year: 2022
  ident: b12
  article-title: Ego4D: Around the world in 3,000 hours of egocentric video
– volume: 245
  start-page: 68
  year: 2022
  end-page: 76
  ident: b13
  article-title: The promise of layer-specific neuroimaging for testing predictive coding theories of psychosis
  publication-title: Schizophrenia Research
– start-page: 1
  year: 2023
  end-page: 15
  ident: b45
  article-title: PreCNet: Next-frame video prediction based on predictive coding
  publication-title: IEEE Transactions on Neural Networks and Learning Systems
– volume: 9
  start-page: 219
  year: 1998
  end-page: 234
  ident: b36
  article-title: Development of localized oriented receptive fields by learning a translation-invariant code for natural images
  publication-title: Network: Computation in Neural Systems
– start-page: 64
  year: 2016
  end-page: 72
  ident: b7
  article-title: Unsupervised learning for physical interaction through video prediction
  publication-title: Proceedings of the 30th international conference on neural information processing systems
– volume: 24
  start-page: 60
  year: 2012
  end-page: 103
  ident: b42
  article-title: Unsupervised learning of generative and discriminative weights encoding elementary image components in a predictive coding model of cortical function
  publication-title: Neural Computation
– volume: 36
  start-page: 181
  year: 2013
  end-page: 204
  ident: b3
  article-title: Whatever next? Predictive brains, situated agents, and the future of cognitive science
  publication-title: Behavioral and Brain Sciences
– year: 2016
  ident: b24
  article-title: Deep predictive coding networks for video prediction and unsupervised learning
– volume: 46
  start-page: 45
  year: 2023
  end-page: 59
  ident: b28
  article-title: Where is the error? Hierarchical predictive coding through dendritic error computation
  publication-title: Trends in Neurosciences
– volume: 7
  year: 2016
  ident: b39
  article-title: Neural elements for predictive coding
  publication-title: Frontiers in Psychology
– volume: 32
  start-page: 1231
  year: 2013
  end-page: 1237
  ident: b10
  article-title: Vision meets robotics: The KITTI dataset
  publication-title: The International Journal of Robotics Research
– volume: 206
  year: 2023
  ident: b19
  article-title: Inconsistent illusory motion in predictive coding deep neural networks
  publication-title: Vision Research
– volume: 2
  start-page: 580
  year: 2011
  end-page: 593
  ident: b15
  article-title: Predictive coding
  publication-title: WIREs Cognitive Science
– volume: 26
  start-page: 371
  year: 2016
  end-page: 376
  ident: b20
  article-title: Selective activation of the deep layers of the human primary visual cortex by top-down feedback
  publication-title: Current Biology
– volume: 19
  year: 2023
  ident: b46
  article-title: Recurrent predictive coding models for associative memory employing covariance learning
  publication-title: PLoS Computational Biology
– volume: 76
  start-page: 695
  year: 2012
  end-page: 711
  ident: b2
  article-title: Canonical microcircuits for predictive coding
  publication-title: Neuron
– year: 2015
  ident: b44
  article-title: Unsupervised learning of video representations using LSTMs
– volume: 225
  year: 2021
  ident: b47
  article-title: Cortical hierarchy, dual counterstream architecture and the importance of top-down generative networks
  publication-title: NeuroImage
– volume: 78
  start-page: 1550
  year: 1990
  end-page: 1560
  ident: b52
  article-title: Backpropagation through time: what it does and how to do it
  publication-title: Proceedings of the IEEE
– volume: 2
  start-page: 210
  year: 2020
  end-page: 219
  ident: b25
  article-title: A neural network trained for prediction mimics diverse features of biological neurons and perception
  publication-title: Nature Machine Intelligence
– volume: 1
  start-page: 1
  year: 1991
  end-page: 47
  ident: b6
  article-title: Distributed hierarchical processing in the primate cerebral cortex
  publication-title: Cerebral Cortex
– volume: 2
  start-page: 79
  year: 1999
  end-page: 87
  ident: b37
  article-title: Predictive coding in the visual cortex: a functional interpretation of some extra-classical receptive-field effects
  publication-title: Nature Neuroscience
– volume: 48
  start-page: 1391
  year: 2008
  end-page: 1408
  ident: b41
  article-title: Predictive coding as a model of biased competition in visual attention
  publication-title: Vision Research
– year: 2022
  ident: b17
  article-title: Predictive coding theories of cortical function
  publication-title: Oxford research encyclopedia of neuroscience
– volume: 360
  start-page: 815
  year: 2005
  end-page: 836
  ident: b9
  article-title: A theory of cortical responses
  publication-title: Philosophical Transactions of the Royal Society, Series B (Biological Sciences)
– year: 2015
  ident: b35
  article-title: Spatio-temporal video autoencoder with differentiable memory
– volume: 195
  start-page: 215
  year: 1968
  end-page: 243
  ident: b16
  article-title: Receptive fields and functional architecture of monkey striate cortex
  publication-title: The Journal of Physiology
– volume: 21
  start-page: 335
  year: 2020
  end-page: 346
  ident: b23
  article-title: Backpropagation and the brain
  publication-title: Nature Reviews. Neuroscience
– year: 2015
  ident: b27
  article-title: Deep multi-scale video prediction beyond mean square error
– volume: Vol. 8
  year: 1995
  ident: b40
  article-title: Unsupervised Pixel-prediction
  publication-title: Advances in neural information processing systems
– volume: 13
  start-page: 600
  year: 2004
  end-page: 612
  ident: b50
  article-title: Image quality assessment: From error visibility to structural similarity
  publication-title: IEEE Transactions on Image Processing
– volume: 15
  year: 2021
  ident: b5
  article-title: Deep gated hebbian predictive coding accounts for emergence of complex neural response properties along the visual cortical hierarchy
  publication-title: Frontiers in Computational Neuroscience
– year: 2012
  ident: b34
  article-title: Prediction as a candidate for learning deep hierarchical models of data
– year: 2015
  ident: b38
  article-title: Convolutional LSTM network: A machine learning approach for precipitation nowcasting
– volume: 32
  start-page: 3665
  year: 2012
  end-page: 3678
  ident: b49
  article-title: A neuronal model of predictive coding accounting for the mismatch negativity
  publication-title: Journal of Neuroscience
– volume: 16
  start-page: 1325
  year: 2003
  end-page: 1352
  ident: b8
  article-title: Learning and inference in the brain
  publication-title: Neural Networks
– start-page: 304
  year: 2009
  end-page: 311
  ident: b4
  article-title: Pedestrian detection: A benchmark
  publication-title: 2009 IEEE conference on computer vision and pattern recognition
– volume: 66
  start-page: 241
  year: 1992
  end-page: 251
  ident: b32
  article-title: On the computational architecture of the neocortex: II the role of cortico-cortical loops
  publication-title: Biological Cybernetics
– volume: 9
  start-page: 345
  year: 2018
  ident: b51
  article-title: Illusory motion reproduced by deep neural networks trained for prediction
  publication-title: Frontiers in Psychology
– volume: Vol. 29
  year: 2016
  ident: b48
  article-title: Generating videos with scene dynamics
  publication-title: Advances in neural information processing systems
– year: 2014
  ident: b33
  article-title: Learning through time in the thalamocortical loops
– volume: 522
  start-page: 225
  year: 2014
  end-page: 259
  ident: b26
  article-title: Anatomy of hierarchy: Feedforward and feedback pathways in macaque visual cortex
  publication-title: Journal of Comparative Neurology
– volume: 39
  start-page: 9410
  year: 2019
  end-page: 9423
  ident: b30
  article-title: Scene representations conveyed by cortical feedback to early visual cortex can be described by line drawings
  publication-title: The Journal of Neuroscience
– volume: 25
  start-page: 2690
  year: 2015
  end-page: 2695
  ident: b31
  article-title: Contextual feedback to superficial layers of V1
  publication-title: Current Biology
– volume: 112
  start-page: 92
  year: 2017
  end-page: 97
  ident: b43
  article-title: A review of predictive coding algorithms
  publication-title: Brain and Cognition
– year: 2020
  ident: b14
  article-title: Hierarchical predictive coding models in a deep-learning framework
– volume: 18
  year: 2020
  ident: b1
  article-title: Prior expectations evoke stimulus-specific activity in the deep layers of the primary visual cortex
  publication-title: PLoS Biology
– year: 2014
  ident: b18
  article-title: Adam: A method for stochastic optimization
– volume: Vol. 28
  year: 2015
  ident: b11
  article-title: Learning to linearize under uncertainty
  publication-title: Advances in neural information processing systems
– volume: 7
  year: 2016
  ident: 10.1016/j.neunet.2025.107246_b39
  article-title: Neural elements for predictive coding
  publication-title: Frontiers in Psychology
  doi: 10.3389/fpsyg.2016.01792
– volume: 20
  start-page: 1434
  issue: 7
  year: 2003
  ident: 10.1016/j.neunet.2025.107246_b21
  article-title: Hierarchical Bayesian inference in the visual cortex
  publication-title: Journal of the Optical Society of America A
  doi: 10.1364/JOSAA.20.001434
– volume: Vol. 29
  year: 2016
  ident: 10.1016/j.neunet.2025.107246_b48
  article-title: Generating videos with scene dynamics
– year: 2020
  ident: 10.1016/j.neunet.2025.107246_b14
– volume: 9
  start-page: 345
  year: 2018
  ident: 10.1016/j.neunet.2025.107246_b51
  article-title: Illusory motion reproduced by deep neural networks trained for prediction
  publication-title: Frontiers in Psychology
  doi: 10.3389/fpsyg.2018.00345
– year: 2014
  ident: 10.1016/j.neunet.2025.107246_b33
– volume: 206
  year: 2023
  ident: 10.1016/j.neunet.2025.107246_b19
  article-title: Inconsistent illusory motion in predictive coding deep neural networks
  publication-title: Vision Research
  doi: 10.1016/j.visres.2023.108195
– volume: 21
  start-page: 335
  issue: 6
  year: 2020
  ident: 10.1016/j.neunet.2025.107246_b23
  article-title: Backpropagation and the brain
  publication-title: Nature Reviews. Neuroscience
  doi: 10.1038/s41583-020-0277-3
– volume: 1
  start-page: 1
  issue: 1
  year: 1991
  ident: 10.1016/j.neunet.2025.107246_b6
  article-title: Distributed hierarchical processing in the primate cerebral cortex
  publication-title: Cerebral Cortex
  doi: 10.1093/cercor/1.1.1
– volume: 26
  start-page: 371
  issue: 3
  year: 2016
  ident: 10.1016/j.neunet.2025.107246_b20
  article-title: Selective activation of the deep layers of the human primary visual cortex by top-down feedback
  publication-title: Current Biology
  doi: 10.1016/j.cub.2015.12.038
– volume: 36
  start-page: 181
  issue: 3
  year: 2013
  ident: 10.1016/j.neunet.2025.107246_b3
  article-title: Whatever next? Predictive brains, situated agents, and the future of cognitive science
  publication-title: Behavioral and Brain Sciences
  doi: 10.1017/S0140525X12000477
– volume: 112
  start-page: 92
  year: 2017
  ident: 10.1016/j.neunet.2025.107246_b43
  article-title: A review of predictive coding algorithms
  publication-title: Brain and Cognition
  doi: 10.1016/j.bandc.2015.11.003
– start-page: 64
  year: 2016
  ident: 10.1016/j.neunet.2025.107246_b7
  article-title: Unsupervised learning for physical interaction through video prediction
– year: 2016
  ident: 10.1016/j.neunet.2025.107246_b24
– volume: 48
  start-page: 1391
  issue: 12
  year: 2008
  ident: 10.1016/j.neunet.2025.107246_b41
  article-title: Predictive coding as a model of biased competition in visual attention
  publication-title: Vision Research
  doi: 10.1016/j.visres.2008.03.009
– volume: 2
  start-page: 210
  issue: 4
  year: 2020
  ident: 10.1016/j.neunet.2025.107246_b25
  article-title: A neural network trained for prediction mimics diverse features of biological neurons and perception
  publication-title: Nature Machine Intelligence
  doi: 10.1038/s42256-020-0170-9
– volume: 522
  start-page: 225
  issue: 1
  year: 2014
  ident: 10.1016/j.neunet.2025.107246_b26
  article-title: Anatomy of hierarchy: Feedforward and feedback pathways in macaque visual cortex
  publication-title: Journal of Comparative Neurology
  doi: 10.1002/cne.23458
– volume: 66
  start-page: 241
  issue: 3
  year: 1992
  ident: 10.1016/j.neunet.2025.107246_b32
  article-title: On the computational architecture of the neocortex: II the role of cortico-cortical loops
  publication-title: Biological Cybernetics
  doi: 10.1007/BF00198477
– year: 2015
  ident: 10.1016/j.neunet.2025.107246_b35
– volume: 195
  start-page: 215
  issue: 1
  year: 1968
  ident: 10.1016/j.neunet.2025.107246_b16
  article-title: Receptive fields and functional architecture of monkey striate cortex
  publication-title: The Journal of Physiology
  doi: 10.1113/jphysiol.1968.sp008455
– volume: Vol. 8
  year: 1995
  ident: 10.1016/j.neunet.2025.107246_b40
  article-title: Unsupervised Pixel-prediction
– volume: 25
  start-page: 2690
  issue: 20
  year: 2015
  ident: 10.1016/j.neunet.2025.107246_b31
  article-title: Contextual feedback to superficial layers of V1
  publication-title: Current Biology
  doi: 10.1016/j.cub.2015.08.057
– volume: 32
  start-page: 1231
  issue: 11
  year: 2013
  ident: 10.1016/j.neunet.2025.107246_b10
  article-title: Vision meets robotics: The KITTI dataset
  publication-title: The International Journal of Robotics Research
  doi: 10.1177/0278364913491297
– volume: Vol. 28
  year: 2015
  ident: 10.1016/j.neunet.2025.107246_b11
  article-title: Learning to linearize under uncertainty
– volume: 225
  year: 2021
  ident: 10.1016/j.neunet.2025.107246_b47
  article-title: Cortical hierarchy, dual counterstream architecture and the importance of top-down generative networks
  publication-title: NeuroImage
  doi: 10.1016/j.neuroimage.2020.117479
– volume: 2
  start-page: 580
  issue: 5
  year: 2011
  ident: 10.1016/j.neunet.2025.107246_b15
  article-title: Predictive coding
  publication-title: WIREs Cognitive Science
  doi: 10.1002/wcs.142
– volume: 76
  start-page: 695
  issue: 4
  year: 2012
  ident: 10.1016/j.neunet.2025.107246_b2
  article-title: Canonical microcircuits for predictive coding
  publication-title: Neuron
  doi: 10.1016/j.neuron.2012.10.038
– volume: 2
  start-page: 79
  issue: 1
  year: 1999
  ident: 10.1016/j.neunet.2025.107246_b37
  article-title: Predictive coding in the visual cortex: a functional interpretation of some extra-classical receptive-field effects
  publication-title: Nature Neuroscience
  doi: 10.1038/4580
– volume: 7
  start-page: 13276
  issue: 1
  year: 2016
  ident: 10.1016/j.neunet.2025.107246_b22
  article-title: Random synaptic feedback weights support error backpropagation for deep learning
  publication-title: Nature Communications
  doi: 10.1038/ncomms13276
– start-page: 304
  year: 2009
  ident: 10.1016/j.neunet.2025.107246_b4
  article-title: Pedestrian detection: A benchmark
– volume: 46
  start-page: 45
  issue: 1
  year: 2023
  ident: 10.1016/j.neunet.2025.107246_b28
  article-title: Where is the error? Hierarchical predictive coding through dendritic error computation
  publication-title: Trends in Neurosciences
  doi: 10.1016/j.tins.2022.09.007
– year: 2014
  ident: 10.1016/j.neunet.2025.107246_b18
– start-page: 1
  year: 2023
  ident: 10.1016/j.neunet.2025.107246_b45
  article-title: PreCNet: Next-frame video prediction based on predictive coding
  publication-title: IEEE Transactions on Neural Networks and Learning Systems
– volume: 19
  issue: 4
  year: 2023
  ident: 10.1016/j.neunet.2025.107246_b46
  article-title: Recurrent predictive coding models for associative memory employing covariance learning
  publication-title: PLoS Computational Biology
  doi: 10.1371/journal.pcbi.1010719
– year: 2015
  ident: 10.1016/j.neunet.2025.107246_b27
– volume: 13
  start-page: 600
  issue: 4
  year: 2004
  ident: 10.1016/j.neunet.2025.107246_b50
  article-title: Image quality assessment: From error visibility to structural similarity
  publication-title: IEEE Transactions on Image Processing
  doi: 10.1109/TIP.2003.819861
– volume: 245
  start-page: 68
  year: 2022
  ident: 10.1016/j.neunet.2025.107246_b13
  article-title: The promise of layer-specific neuroimaging for testing predictive coding theories of psychosis
  publication-title: Schizophrenia Research
  doi: 10.1016/j.schres.2020.10.009
– volume: 360
  start-page: 815
  issue: 1456
  year: 2005
  ident: 10.1016/j.neunet.2025.107246_b9
  article-title: A theory of cortical responses
  publication-title: Philosophical Transactions of the Royal Society, Series B (Biological Sciences)
– year: 2022
  ident: 10.1016/j.neunet.2025.107246_b17
  article-title: Predictive coding theories of cortical function
– year: 2022
  ident: 10.1016/j.neunet.2025.107246_b12
– volume: 15
  year: 2021
  ident: 10.1016/j.neunet.2025.107246_b5
  article-title: Deep gated hebbian predictive coding accounts for emergence of complex neural response properties along the visual cortical hierarchy
  publication-title: Frontiers in Computational Neuroscience
  doi: 10.3389/fncom.2021.666131
– volume: 78
  start-page: 1550
  issue: 10
  year: 1990
  ident: 10.1016/j.neunet.2025.107246_b52
  article-title: Backpropagation through time: what it does and how to do it
  publication-title: Proceedings of the IEEE
  doi: 10.1109/5.58337
– volume: 32
  start-page: 3665
  issue: 11
  year: 2012
  ident: 10.1016/j.neunet.2025.107246_b49
  article-title: A neuronal model of predictive coding accounting for the mismatch negativity
  publication-title: Journal of Neuroscience
  doi: 10.1523/JNEUROSCI.5003-11.2012
– volume: 16
  start-page: 1325
  issue: 9
  year: 2003
  ident: 10.1016/j.neunet.2025.107246_b8
  article-title: Learning and inference in the brain
  publication-title: Neural Networks
  doi: 10.1016/j.neunet.2003.06.005
– volume: 39
  start-page: 9410
  issue: 47
  year: 2019
  ident: 10.1016/j.neunet.2025.107246_b30
  article-title: Scene representations conveyed by cortical feedback to early visual cortex can be described by line drawings
  publication-title: The Journal of Neuroscience
  doi: 10.1523/JNEUROSCI.0852-19.2019
– volume: 18
  issue: 12
  year: 2020
  ident: 10.1016/j.neunet.2025.107246_b1
  article-title: Prior expectations evoke stimulus-specific activity in the deep layers of the primary visual cortex
  publication-title: PLoS Biology
  doi: 10.1371/journal.pbio.3001023
– year: 2012
  ident: 10.1016/j.neunet.2025.107246_b34
– volume: 9
  start-page: 219
  issue: 2
  year: 1998
  ident: 10.1016/j.neunet.2025.107246_b36
  article-title: Development of localized oriented receptive fields by learning a translation-invariant code for natural images
  publication-title: Network: Computation in Neural Systems
– year: 2015
  ident: 10.1016/j.neunet.2025.107246_b38
– year: 2015
  ident: 10.1016/j.neunet.2025.107246_b44
– year: 2023
  ident: 10.1016/j.neunet.2025.107246_b29
  article-title: Predictive coding networks for temporal prediction
  publication-title: Neuroscience
– volume: 24
  start-page: 60
  issue: 1
  year: 2012
  ident: 10.1016/j.neunet.2025.107246_b42
  article-title: Unsupervised learning of generative and discriminative weights encoding elementary image components in a predictive coding model of cortical function
  publication-title: Neural Computation
  doi: 10.1162/NECO_a_00222
SSID ssj0006843
Score 2.455959
Snippet Hierarchical predictive coding (hPC) provides a compelling framework for understanding how the cortex predicts future sensory inputs by minimizing prediction...
SourceID proquest
pubmed
crossref
elsevier
SourceType Aggregation Database
Index Database
Publisher
StartPage 107246
SubjectTerms Deep Learning
Humans
Models, Neurological
Neural Networks, Computer
Predictive coding
Predictive processing
Temporal prediction
Unsupervised learning
Unsupervised Machine Learning
Title PrediRep: Modeling hierarchical predictive coding with an unsupervised deep learning network
URI https://dx.doi.org/10.1016/j.neunet.2025.107246
https://www.ncbi.nlm.nih.gov/pubmed/39946763
https://www.proquest.com/docview/3166765049
Volume 185
WOSCitedRecordID wos001427277300001&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: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1879-2782
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0006843
  issn: 0893-6080
  databaseCode: AIEXJ
  dateStart: 19950101
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lb9QwELbolgMX3o_lURkJcamy2sRxnHCr0FYUVUsPWykHJMvr2IUevGGzQf35jGM7CSoV5cAlirJ5refz5Jvx2B9C72hasTgVcaQZIxCgSBkBlGmkLHulwPHZXHZiE2y5zMuyOPMFmU0nJ8CMya-uivq_mhqOgbHt1Nl_MHd_UzgA-2B02ILZYXsrw59t4XsEvNrG-lbprJtvbhWvuzEDa5LantH5uUO5qfpsLPT01jRtbb1HAzy0UqoOqhIXh8bVi4_JrF3YA27nf2kGZ9Z8cxrNJxCKw-4oE9sJ7vo5Qq4CzC3OD0buctF2uMnXC_tUREKHwj_vPXNWRAlzakIz9YdjvculI6cJEWji8pDX_LlLLVzOjGrhz8zsQ2fD6b8vn738wo_PT0_5alGu3tc_IqssZkfgvczKHtpPGC3yCdo_OlmUn_vvdZa7aRjhRcMEy64K8PqDbyIwNwUoHVFZPUT3fYSBjxwyHqE7yjxGD4J6B_bO_An6GoDyAQeY4DFM8AAT7GCCLUywMHgME2xhggNMsAfDU3R-vFh9_BR5rY1IEsp2EVNzLSpRFFbDU2oZJ2utmaTrbE1jqeZU6nU61wkR0HkTKSUwT0alotWcMC1S8gxNzMaoFwjLIiVExFoxVaUa6A7LFNGxYFlGC8HIFEWhAXntllThodbwkrsG57bBuWvwKWKhlbmnhY7ucUDJX658G4zCwWvaoTBh1KZtOIltbTeF8HiKnjtr9e8ClB3YQ0Ze3uLqV-je0BFeo8lu26o36K78ufvebA_QHivzA4-3Xx9OmpI
linkProvider Elsevier
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=PrediRep%3A+Modeling+hierarchical+predictive+coding+with+an+unsupervised+deep+learning+network&rft.jtitle=Neural+networks&rft.au=Hashim%2C+Ibrahim+C&rft.au=Senden%2C+Mario&rft.au=Goebel%2C+Rainer&rft.date=2025-05-01&rft.issn=1879-2782&rft.eissn=1879-2782&rft.volume=185&rft.spage=107246&rft_id=info:doi/10.1016%2Fj.neunet.2025.107246&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0893-6080&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0893-6080&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0893-6080&client=summon