Learning invariance from transformation sequences
Abstract
The visual system can reliably identify objects even when the retinal image is transformed considerably by commonly occurring changes in the environment. A local learning rule is proposed, which allows a network to learn to generalize across such transformations. During the learning phase, the network is exposed to temporal sequences of patterns undergoing the transformation. An application of the algorithm is presented in which the network learns invariance to shift in retinal position. Such a principle may be involved in the development of the characteristic shift invariance property of complex cells in the primary visual cortex, and also in the development of more complicated invariance properties of neurons in higher visual areas.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
This idea of continuity is used in a great number of artificial neural networks and related models.
“This idea of continuity is used in a great number of artificial neural networks and related models (Wiskott and Sejnowski, 2002; Földiák, 2008; Mobahi et al., 2009).”
From Toward an integration of deep learning and neuroscience (2016) · cites this paper as Földiák, 2008
Cites
Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex
Hubel, D. H., & Wiesel, T. N. (1962) · The Journal of Physiology · doi:10.1113/jphysiol.1962.sp006837 · Cited by 13,680 · see citing works
Backpropagation Applied to Handwritten Zip Code Recognition
LeCun, Y., Boser, B., Denker, J. S., Henderson, D., Howard, R. E., Hubbard, W., & Jackel, L. D. (1989) · Neural Computation · doi:10.1162/neco.1989.1.4.541 · Cited by 11,891 · see citing works
Fukushima, K. (1980) · Biological Cybernetics · doi:10.1007/bf00344251 · Cited by 5,759 · see citing works
Toward a modern theory of adaptive networks: Expectation and prediction.
Sutton, R. S., & Barto, A. G. (1981) · Psychological Review · doi:10.1037/0033-295x.88.2.135 · Cited by 1,450 · see citing works
Visual neurones responsive to faces in the monkey temporal cortex
Perrett, D. I., Rolls, E. T., & Caan, W. (1982) · Experimental Brain Research · doi:10.1007/bf00239352 · Cited by 1,386 · see citing works
Feature Discovery by Competitive Learning*
Rumelhart, D. E., & Zipser, D. (1985) · Cognitive Science · doi:10.1207/s15516709cog0901_5 · Cited by 1,379 · see citing works
Forming sparse representations by local anti-Hebbian learning
Földiák, P. (1990) · Biological Cybernetics · doi:10.1007/bf02331346 · Cited by 546 · see citing works
Adaptation and decorrelation in the cortex
Barlow, H. B., & Földiák, P. (1989) · Addison-Wesley Longman Publishing Co., Inc. eBooks · Cited by 263 · see citing works
Learning translation invariant recognition in a massively parallel networks
Hinton, G. E. (1987) · Lecture notes in computer science · doi:10.1007/3-540-17943-7_117 · Cited by 177 · see citing works
A complex-cell receptive-field model
Spitzer, H., & Hochstein, S. (1985) · Journal of Neurophysiology · doi:10.1152/jn.1985.53.5.1266 · Cited by 144 · see citing works
The Hedonistic Neuron: A Theory of Memory, Learning and Intelligence
Edwards, J. S. (1983) · Journal of the Operational Research Society · doi:10.1057/jors.1983.154 · Cited by 137 · see citing works
The Neural Basis of Stimulus Equivalence Across Retinal Translation
Gross, C. G., & Mishkin, M. M. (1977) · Elsevier eBooks · doi:10.1016/b978-0-12-325750-5.50014-2 · Cited by 114 · see citing works
Removing Time Variation with the Anti-Hebbian Differential Synapse
Mitchison, G. (1991) · Neural Computation · doi:10.1162/neco.1991.3.3.312 · Cited by 60 · see citing works
Self-organization of position- and deformation-tolerant neural representations
Webber, C. J. S. (1991) · Network Computation in Neural Systems · doi:10.1088/0954-898x_2_1_003 · Cited by 10 · see citing works
Flagged first, then by citations
1 of the 15 references couldn’t be matched to a known paper.
Cited by
Sorted by citations
Hierarchical models of object recognition in cortex
Riesenhuber, M., & Poggio, T. (1999) · Nature Neuroscience · doi:10.1038/14819 · Cited by 3,309 · see citing works
Unsupervised Visual Representation Learning by Context Prediction
Doersch, C., Gupta, A., & Efros, A. A. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.167 · Cited by 2,629 · see citing works
“This kind of temporal coherence has a long history in visual learning literature [18, 56], and contemporaneous work shows strong improvements on modern detection datasets [54].”
BackgroundHow does the brain solve visual object recognition?
DiCarlo, J. J., Zoccolan, D., & Rust, N. C. (2012) · Neuron · doi:10.1016/j.neuron.2012.01.010 · Cited by 1,790 · see citing works
“More specifically, because each object’s identity is temporally stable, different retinal images of the same object tend to be temporally contiguous (Fazl et al., 2009; Foldiak, 1991; Stryker, 1992;Wallis and Rolls, 1997; Wiskott and Sejnowski, 2002).”
Deep supervised, but not unsupervised, models may explain IT cortical representation
Khaligh-Razavi, S.-M., & Kriegeskorte, N. (2014) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1003915 · Cited by 1,204 · see citing works
Slow Feature Analysis: Unsupervised Learning of Invariances
Wiskott, L., & B, T. J. S. A. (2002) · Neural Computation · doi:10.1162/089976602317318938 · Cited by 1,351 · see citing works
Toward Causal Representation Learning
Scholkopf, B., Locatello, F., Bauer, S., Ke, N. R., Kalchbrenner, N., Goyal, A., & Bengio, Y. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2021.3058954 · Cited by 1,170 · see citing works
“The hypothesis is also related to ideas of looking for features that vary slowly [70], [270].”
BackgroundUnsupervised Learning of Visual Representations Using Videos
Wang, X., & Gupta, A. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.320 · Cited by 1,005 · see citing works
“Therefore, researchers have started focusing on learning feature representations using videos [11, 53, 27, 43, 56, 16, 48, 34, 45].”
BackgroundUntangling invariant object recognition
DiCarlo, J. J., & Cox, D. (2007) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2007.06.010 · Cited by 948 · see citing works
Olshausen, B., Anderson, C., & Van Essen, D. (1993) · The Journal of Neuroscience · doi:10.1523/jneurosci.13-11-04700.1993 · Cited by 908 · see citing works
Shuffle and Learn: Unsupervised Learning Using Temporal Order Verification
Misra, I., Zitnick, C. L., & Hebert, M. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46448-0_32 · Cited by 779 · see citing works
Matsugu, M., Mori, K., Mitari, Y., & Kaneda, Y. (2003) · Neural Networks · doi:10.1016/s0893-6080(03)00115-1 · Cited by 715 · see citing works
“Most algorithms, so far, have addressed only a part of these problems (Földiák, 1991; Wallis & Rolls, 1997).”
BackgroundDeep Predictive Coding Networks for Video Prediction and Unsupervised LearningPreprint
Lotter, W. E., Kreiman, G., & Cox, D. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.08104 · Cited by 805 · see citing works
Face Recognition by Humans: Nineteen Results All Computer Vision Researchers Should Know About
Sinha, P., Balas, B., Ostrovsky, Y., & Russell, R. (2006) · Proceedings of the IEEE · doi:10.1109/jproc.2006.884093 · Cited by 674 · see citing works
“Exploring image sequences using mechanisms that make explicit temporal associations [ 22 ] may be a powerful means for view generalization.”
BackgroundRiesenhuber, M., & Poggio, T. (2000) · Nature Neuroscience · doi:10.1038/81479 · Cited by 658 · see citing works
Multi-task Self-Supervised Visual Learning
Doersch, C., & Zisserman, A. (2017) · IEEE International Conference on Computer Vision · doi:10.1109/iccv.2017.226 · Cited by 682 · see citing works
“Another is to note that objects generally appear across many frames: thus, we can train features to remain invariant as a video progresses [11, 24, 42, 43, 47].”
BackgroundComputational Neuroscience of Vision
Rolls, E. T., & Deco, G. (2001) · doi:10.1093/acprof:oso/9780198524885.001.0001 · Cited by 551 · see citing works
Unsupervised Learning of Visual Features through Spike Timing Dependent Plasticity
Masquelier, T., & Thorpe, S. J. (2007) · PLoS Computational Biology · doi:10.1371/journal.pcbi.0030031 · Cited by 549 · see citing works
INVARIANT FACE AND OBJECT RECOGNITION IN THE VISUAL SYSTEM
Wallis, G. M., & Rolls, E. T. (1997) · Progress in Neurobiology · doi:10.1016/s0301-0082(96)00054-8 · Cited by 539 · see citing works
Hyvärinen, A., & Hoyer, P. O. (2000) · Neural Computation · doi:10.1162/089976600300015312 · Cited by 533 · see citing works
“Similar developments were given by F¨oldi´ak (1991).”
Neuronal Mechanisms of Object Recognition
Tanaka, K. (1993) · Science · doi:10.1126/science.8235589 · Cited by 510 · see citing works
View-invariant representations of familiar objects by neurons in the inferior temporal visual cortex
Booth, M. J. (1998) · Cerebral Cortex · doi:10.1093/cercor/8.6.510 · Cited by 464 · see citing works
“If one feature temporally follows another feature, these features may become associated together ( Foldiak, 1991; Rolls, 1992).”
BackgroundLearning Correspondence From the Cycle-Consistency of Time
Wang, X., Jabri, A., & Efros, A. A. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00267 · Cited by 452 · see citing works
“Computational approaches for unsupervised learning have sought to leverage this continuity, such as continuous transformation learning [13, 70], “slow” feature learning [76, 91, 25] and information maximization between neighbouring patches in time [66].”
BackgroundRegularizing and Optimizing LSTM Language Models
Merity, S., Keskar, N. S., & Socher, R. (2017) · arXiv · doi:10.48550/arxiv.1708.02182 · Cited by 697 · see citing works
“TAR falls under the broad category of slowness regularizers (Hinton, 1989; Földiák, 1991; Luciw & Schmidhuber, 2012; Jonschkowski & Brock, 2015) which penalize the model from producing large changes in the hidden state.”
BackgroundTowards a mathematical theory of cortical micro-circuits
George, D., & Hawkins, J. (2009) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1000532 · Cited by 404 · see citing works
Processing speed in the cerebral cortex and the neurophysiology of visual masking
Rolls, E. T., & Tovée, M. J. (1994) · Proceedings of the Royal Society B Biological Sciences · doi:10.1098/rspb.1994.0087 · Cited by 398 · see citing works
Prediction as a candidate for learning deep hierarchical models of data
Palm, R. (2012) · Cited by 378 · see citing works
Rolls, E. T. (1992) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.1992.0002 · Cited by 374 · see citing works
“Foldiak (1991) has proposed computing an average activation of the postsynaptic neuron to assist with the same problem, but the idea here is that the temporal properties of the biologically implemented learning mechanism are such that it is”
BackgroundMethodA computational theory of hippocampal function, and tests of the theory: New developments
Kesner, R. P., & Rolls, E. T. (2015) · Neuroscience & Biobehavioral Reviews · doi:10.1016/j.neubiorev.2014.11.009 · Cited by 347 · see citing works
In search of common foundations for cortical computation
Phillips, W. A., & Singer, W. (1997) · Behavioral and Brain Sciences · doi:10.1017/s0140525x9700160x · Cited by 338 · see citing works
A Neurodynamical cortical model of visual attention and invariant object recognition
Deco, G., & Rolls, E. T. (2003) · Vision Research · doi:10.1016/j.visres.2003.09.037 · Cited by 316 · see citing works
“By presenting consistent sequences of transforming objects the cells in the network can learn to respond to the same object when it is presented in different locations, as described by F€ o oldi a ak (1991), Rolls (1992) and Wallis and Rolls (1997).”
BackgroundRolls, E. T. (2000) · Neuron · doi:10.1016/s0896-6273(00)00030-1 · Cited by 303 · see citing works
Wachsmuth, E., Oram, M. W., & Perrett, D. I. (1994) · Cerebral Cortex · doi:10.1093/cercor/4.5.509 · Cited by 278 · see citing works
Neuroconstructivism - IHow the brain constructs cognition
Mareschal, D., Johnson, M. H., Sirois, S., Spratling, M., Thomas, M. S. C., & Westermann, G. (2007) · Oxford University Press eBooks · doi:10.1093/acprof:oso/9780198529910.001.0001 · Cited by 271 · see citing works
Unsupervised Natural Experience Rapidly Alters Invariant Object Representation in Visual Cortex
Li, N., & DiCarlo, J. J. (2008) · Science · doi:10.1126/science.1160028 · Cited by 263 · see citing works
Perrett, D. I., Oram, M. W., & Ashbridge, E. (1998) · Cognition · doi:10.1016/s0010-0277(98)00015-8 · Cited by 255 · see citing works
“Competitive learning mechanisms (e.g. Földiák, 1991) coupled with visual experience may account both for cell selectivity for objects that are frequently seen and for the cell sensitivity to the viewing conditions in which the objects are seen.”
BackgroundSerre, T., Kouh, M., Cadieu, C. F., Knoblich, U., Kreiman, G., & Poggio, T. (2005) · Defense Technical Information Center (DTIC) · Cited by 245 · see citing works
Learning to Represent Spatial Transformations with Factored Higher-Order Boltzmann Machines
Memisevic, R., & Hinton, G. E. (2010) · Neural Computation · doi:10.1162/neco.2010.01-09-953 · Cited by 227 · see citing works
“An early approach to learning image transformations in a biologically plausible way, using temporal coherence, is described in Foldiak (1991). More recently, several multilinear and spatiotemporal filtering models were suggested for this task (e.”
MethodInfluentialA Model of V4 Shape Selectivity and Invariance
Cadieu, C. F., Kouh, M., Pasupathy, A., Connor, C. E., Riesenhuber, M., & Poggio, T. (2007) · Journal of Neurophysiology · doi:10.1152/jn.01265.2006 · Cited by 226 · see citing works
“The invariance may also be learned in a biophysically plausible way (e.g., Foldiak 1991; Wallis 1996; Wiskott and Sejnowski 2002), during a developmental period, from natural viewing experiences , such as watching a temporal sequence of moving objects .”
Hierarchical Neural Networks for Image Interpretation
Behnke, S. (2003) · Lecture notes in computer science · doi:10.1007/b11963 · Cited by 225 · see citing works
Object recognition and segmentation by a fragment-based hierarchy
Ullman, S. (2006) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2006.11.009 · Cited by 225 · see citing works
A hierarchical bayesian model of invariant pattern recognition in the visual cortex
Dileep George, A., & Hawkins, J. (2006) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2005.1556155 · Cited by 224 · see citing works
“We believe that continuity of time is the cue that brain uses to solve the invariance problem [5], [18].”
BackgroundA quantitative theory of immediate visual recognition
Serre, T., Kreiman, G., Kouh, M., Cadieu, C. F., Knoblich, U., & Poggio, T. (2007) · Progress in brain research · doi:10.1016/s0079-6123(06)65004-8 · Cited by 217 · see citing works
“Stereo (in addition to motion) is likely to play an important role in the learning of invariances such as position and size invariance via a correlation-based rule such as the trace rule (Földiák, 1991).”
BackgroundWallis, G. M., & Bülthoff, H. (1999) · Trends in Cognitive Sciences · doi:10.1016/s1364-6613(98)01261-3 · Cited by 219 · see citing works
Slowness and Sparseness Lead to Place, Head-Direction, and Spatial-View Cells
Franzius, M., Sprekeler, H., & Wiskott, L. (2007) · PLoS Computational Biology · doi:10.1371/journal.pcbi.0030166 · Cited by 218 · see citing works
Neurophysiology of shape processing
Perrett, D. I., & Oram, M. W. (1993) · Image and Vision Computing · doi:10.1016/0262-8856(93)90011-5 · Cited by 213 · see citing works
Effects of temporal association on recognition memory
Wallis, G. M., & Bülthoff, H. (2001) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.071028598 · Cited by 212 · see citing works
DiCarlo, J. J., & Maunsell, J. H. R. (2003) · Journal of Neurophysiology · doi:10.1152/jn.00358.2002 · Cited by 207 · see citing works
“, Poggio 1990; Wallis and Rolls 1997), including the limited position tolerance of complex cells in V1 (Foldiak 1991).”
BackgroundRolls, E. T., & Milward, T. (2000) · Neural Computation · doi:10.1162/089976600300014845 · Cited by 204 · see citing works
Unsupervised Representation Learning With Long-Term Dynamics for Skeleton Based Action Recognition
Zheng, N., Wen, J., Liu, R., Long, L., Dai, J., & Gong, Z. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11853 · Cited by 201 · see citing works
“Wiskott and Sejnowski proposed the slow feature analysis framework for exploiting temporal structure in sequences and attempted to learn slowly-varying representations (Földiák 2008).”
BackgroundInfluentialSpatio-temporal Self-Supervised Representation Learning for 3D Point Clouds
Huang, S., Xie, Y., Zhu, S. C., & Zhu, Y. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00647 · Cited by 195 · see citing works
“To learn the invariant representation [12, 68], we explore the inextricably spatio-temporal contextual cues embedded in 3D point clouds.”
BackgroundLearning Classification with Unlabeled Data
de Sa, V. R. (1993) · Cited by 189 · see citing works
Deep neural networks in computational neuroscience
Kietzmann, T. C., McClure, P., & Kriegeskorte, N. (2019) · Oxford Research Encyclopedia of Neuroscience · doi:10.1093/acrefore/9780190264086.013.46 · Cited by 186 · see citing works
Cerebral cortex : principles of operation
Rolls, E. T. (2016) · doi:10.82444/warw.33712699 · Cited by 187 · see citing works
Unsupervised Feature Extraction by Time-Contrastive Learning and Nonlinear ICAPreprint
Hyvärinen, A., & Morioka, H. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.06336 · Cited by 176 · see citing works
“The idea is to find features which change as slowly as possible, originally proposed in [6] for learning invariant features.”
BackgroundCorrelations of cortical Hebbian reverberations: theory versus experiment
Amit, D., Brunel, N., & Tsodyks, M. V. (1994) · Journal of Neuroscience · doi:10.1523/jneurosci.14-11-06435.1994 · Cited by 176 · see citing works
Pinto, N., Doukhan, D., DiCarlo, J. J., & Cox, D. D. (2009) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1000579 · Cited by 248 · see citing works
Invariant Visual Object and Face Recognition: Neural and Computational Bases, and a Model, VisNet
Rolls, E. T. (2012) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2012.00035 · Cited by 169 · see citing works
“Földiák (1991) has proposed computing an average activation of the post-synaptic neuron to assist with the same problem.”
BackgroundPolk, T. A., & Farah, M. J. (1998) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.95.3.847 · Cited by 170 · see citing works
Learning optimized features for hierarchical models of invariant object recognition.
Wersing, H., & Körner, E. (2003) · Neural Computation · doi:10.1162/089976603321891800 · Cited by 169 · see citing works
'Breaking' position-invariant object recognition
Cox, D., Meier, P. M., Oertelt, N., & DiCarlo, J. J. (2005) · Nature Neuroscience · doi:10.1038/nn1519 · Cited by 158 · see citing works
A Canonical Neural Circuit for Cortical Nonlinear Operations
Kouh, M., & Poggio, T. (2008) · Neural Computation · doi:10.1162/neco.2008.02-07-466 · Cited by 163 · see citing works
“We also note that the invariance property can be learned by the similar Hebbian mechanism (Bartlett & Sejnowski, 1998; Földiák, 1991).”
BackgroundDenoising Source SeparationPreprint
Särelä, J., & Harri Valpola, B. (2005) · CogPrints (University of Southampton) · Cited by 162 · see citing works
Karklin, Y., & Lewicki, M. S. (2005) · Neural Computation · doi:10.1162/0899766053011474 · Cited by 155 · see citing works
“…of natural signals change slowly across time or space and that this spatial and temporal coherence can be used to extract higher-order structure from the data ( Foldiak, 1991; Kayser, Einhäuser, D ¨ ummer, K ¨ onig, & K ¨ ording, 2001; Wiskott & Sejnowski , 2002; Hurri & Hyvärinen, 2003 ).”
MethodNeural Representations for Object Perception: Structure, Category, and Adaptive Coding
Kourtzi, Z., & Connor, C. E. (2011) · Annual Review of Neuroscience · doi:10.1146/annurev-neuro-060909-153218 · Cited by 146 · see citing works
“Experience and training may be a solution to this problem (Foldiak 1991, Wallis & Rolls 1997, Ullman & Soloviev 1999, Wallis & Bulthoff 2001) by enhancing the sparseness and clustering of the neural code. fMRI studies show that at the level of large neural populations training results in…”
BackgroundThe representation of information about faces in the temporal and frontal lobes
Rolls, E. T. (2006) · Neuropsychologia · doi:10.1016/j.neuropsychologia.2006.04.019 · Cited by 152 · see citing works
A model of the ventral visual system based on temporal stability and local memory
Wyss, R., König, P., & Verschure, P. F. M. J. (2006) · PLoS Biology · doi:10.1371/journal.pbio.0040120 · Cited by 145 · see citing works
Self-Organized Formation of Various Invariant-Feature Filters in the Adaptive-Subspace SOM
Kohonen, T., Samuel Kaski, A., & Lappalainen, H. (1997) · Neural Computation · doi:10.1162/neco.1997.9.6.1321 · Cited by 145 · see citing works
“One should mention a couple of recent works (Földiák 1991; Wallis et al. 1993) in which neural units learn sequences of simple patterns.”
BackgroundFigure-ground organization and object recognition processes: An interactive account.
Vecera, S. P., & O’Reilly, R. C. (1998) · Journal of Experimental Psychology Human Perception & Performance · doi:10.1037//0096-1523.24.2.441 · Cited by 136 · see citing works
Modeling visual recognition from neurobiological constraints
Oram, M. W., & Perrett, D. I. (1994) · Neural Networks · doi:10.1016/s0893-6080(05)80153-4 · Cited by 134 · see citing works
Li, N., & DiCarlo, J. J. (2010) · Neuron · doi:10.1016/j.neuron.2010.08.029 · Cited by 130 · see citing works
“…contiguity of object features during natural visual experience can instruct the learning of tolerance, potentially in an unsupervised manner (Foldiak, 1991; Masquelier et al., 2007; Masquelier and Thorpe, 2007; Sprekeler et al., 2007; Stryker, 1991; Wiskott and Sejnowski, 2002; Wyss et al.,…”
InfluentialBio-inspired unsupervised learning of visual features leads to robust invariant object recognition
Kheradpisheh, S. R., Ganjtabesh, ., & Masquelier, T. (2016) · Neurocomputing · doi:10.1016/j.neucom.2016.04.029 · Cited by 129 · see citing works
“A prominent example is the trace learning rule [64], suggesting that the invariant object representation in ventral visual system is instructed by the implicit temporal contiguity of vision.”
BackgroundInvariant Visual Responses From Attentional Gain Fields
Salinas, E., & Abbott, L. F. (1997) · Journal of Neurophysiology · doi:10.1152/jn.1997.77.6.3267 · Cited by 129 · see citing works
Towards cortex sized artificial neural systems
Johansson, C., & Lansner, A. (2006) · Neural Networks · doi:10.1016/j.neunet.2006.05.029 · Cited by 128 · see citing works
A cocktail party with a cortical twist: How cortical mechanisms contribute to sound segregation
Elhilali, M., & Shamma, S. (2008) · The Journal of the Acoustical Society of America · doi:10.1121/1.3001672 · Cited by 125 · see citing works
How are complex cell properties adapted to the statistics of natural stimuli?
Körding, K. P., Kayser, C., Einhäuser, W., & König, P. (2004) · Journal of Neurophysiology · doi:10.1152/jn.00149.2003 · Cited by 121 · see citing works
“These studies have successfully applied this criterion to the representations of artificial stimuli such as moving bars to establish that such a mechanism could lead to complex-type neurons (Földiak 1991; Wiskott and Sejnowski 2002).”
BackgroundInfluentialShape Tuning in Macaque Inferior Temporal Cortex
Kayaert, G., Biederman, I., & Vogels, R. (2003) · Journal of Neuroscience · doi:10.1523/jneurosci.23-07-03016.2003 · Cited by 121 · see citing works
“Indeed, single IT neurons may learn to respond similarly to temporally contiguous images (Foldiak, 1991; Wallis, 1998) (for physiological evidence, see Miyashita, 1988).”
BackgroundToward an integration of deep learning and neuroscience
Marblestone, A. H., Wayne, G., & Kording, K. P. (2016) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2016.00094 · Cited by 119 · see citing works
“This idea of continuity is used in a great number of artificial neural networks and related models (Wiskott and Sejnowski, 2002; Földiák, 2008; Mobahi et al., 2009).”
MethodPrécis ofNeuroconstructivism: How the Brain Constructs Cognition
Sirois, S., Spratling, M., Thomas, M. S. C., Westermann, G., Mareschal, D., & Johnson, M. H. (2008) · Behavioral and Brain Sciences · doi:10.1017/s0140525x0800407x · Cited by 120 · see citing works
“This is achieved through an interactive process involving feedforward and feedback connections between the different cortical areas involved in object perception (Driver et al. 2001).”
BackgroundMethodBrain mechanisms for invariant visual recognition and learning
Rolls, E. T. (1994) · Behavioural Processes · doi:10.1016/0376-6357(94)90062-0 · Cited by 116 · see citing works
Video Jigsaw: Unsupervised Learning of Spatiotemporal Context for Video Action Recognition
Ahsan, U., Madhok, R., & Essa, I. A. (2019) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2019.00025 · Cited by 113 · see citing works
Human eye-head co-ordination in natural exploration
Einhäuser, W., Schumann, F., Bardins, S., Bartl, K., Böning, G., Schneider, E., & Peter König, B. (2007) · Network Computation in Neural Systems · doi:10.1080/09548980701671094 · Cited by 114 · see citing works
A multiscale dynamic routing circuit for forming size- and position-invariant object representations
Olshausen, B. A., Anderson, C. H., & Essen, D. C. V. (1995) · Journal of Computational Neuroscience · doi:10.1007/bf00962707 · Cited by 113 · see citing works
“One promising approach has been described by Foldiak (1991) , who has shown how a complex cell can learn translation invariance by assuming that the presence of an object (or feature) is stable over time.”
BackgroundFace Image Analysis by Unsupervised Learning
Bartlett, M. S. (2001) · Springer eBooks · doi:10.1007/978-1-4615-1637-8 · Cited by 109 · see citing works
“Földíak (Földiák, 1991) demonstrated that Hebbian learning can capture temporal relationships in a feedforward system when the output unit activities undergo temporal smoothing.”
BackgroundMethodInfluentialA Matching Advantage for Dynamic Human Faces
Thornton, I. M., & Kourtzi, Z. (2002) · Perception · doi:10.1068/p3300 · Cited by 110 · see citing works
“Other recent computational (eg Edelman and Weinshall 1991; Foldiak 1991; Wallis and Rolls 1997) and behavioural (eg Sinha and Poggio 1996; Stone 1998; Stone and Harper 1999; Wallis 1998) studies also indicate that temporal correlations, such as those associated with motion, can affect the long-term…”
BackgroundWeakly-Supervised Disentanglement Without CompromisesPreprint
Locatello, F., Poole, B., Rätsch, G., Schölkopf, B., Bachem, O., & Tschannen, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.02886 · Cited by 117 · see citing works
“More generally this observation model applies to many natural scenes with moving objects (Földiák, 1991).”
BackgroundA Developmental Approach to Machine Learning?
Smith, L. B., & Slone, L. K. (2017) · Frontiers in Psychology · doi:10.3389/fpsyg.2017.02124 · Cited by 107 · see citing works
“…extended time, the retinal information with respect to that object will necessarily and continuously change, revealing relevant transformations and the invariances for recognition that may be extendable to recognizing novel things (Földiák, 1991; Wiskott and Sejnowski, 2002; Li and DiCarlo, 2008).”
BackgroundCan inhibition resolve retrieval competition through the control of spreading activation?
Saunders, J., & MacLeod, M. D. (2006) · Memory & Cognition · doi:10.3758/bf03193409 · Cited by 106 · see citing works
“…can still be considered important, in the sense that it keeps categories separate and allows a winner-takes-all strategy to be implemented; that is, when one category is activated, it lessens the likelihood of other categories being similarly activated (see Földiák, 1991; Oram & Földiák, 1996).”
BackgroundA Maximum-Likelihood Interpretation for Slow Feature Analysis
Turner, R. E., & Sahani, M. (2007) · Neural Computation · doi:10.1162/neco.2007.19.4.1022 · Cited by 104 · see citing works
“Observations of this sort motivate temporal constancy, or slowness, as a useful learning principle to extract meaningful higher-level descriptions of sensory data (Hinton, 1989; Földiák, 1991; Mitchison, 1991; Stone, 1996).”
BackgroundSalinas, E., & Abbott, L. F. (2001) · Progress in brain research · doi:10.1016/s0079-6123(01)30012-2 · Cited by 99 · see citing works
Aggregation via Separation: Boosting Facial Landmark Detector With Semi-Supervised Style Translation
Qian, S., Sun, K., Wu, W., Qian, C., & Jia, J. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.01025 · Cited by 99 · see citing works
“Visual invariance of the same instance could be captured in a consecutive sequence of video frames [20, 67, 39, 85, 59, 58, 65].”
BackgroundLearning and neural plasticity in visual object recognition
Kourtzi, Z., & DiCarlo, J. J. (2006) · Current Opinion in Neurobiology · doi:10.1016/j.conb.2006.03.012 · Cited by 100 · see citing works
“At a theoretical level, there is growing appreciation of the potentially powerful role of learning in establishing robust representations crucial for object recognition [7–12].”
BackgroundSimple-Cell-Like Receptive Fields Maximize Temporal Coherence in Natural Video
Hurri, J., & Hyvärinen, A. (2003) · Neural Computation · doi:10.1162/089976603321192121 · Cited by 99 · see citing works
“F(cid:127)oldi(cid:19)ak was one of the (cid:12)rst authors to suggest the usefulness of temporal coherence in computational neuroscience (F(cid:127)oldi(cid:19)ak, 1991).”
BackgroundInvariant Object Recognition in the Visual System with Novel Views of 3D Objects
Stringer, S. M., & Rolls, E. T. (2002) · Neural Computation · doi:10.1162/089976602760407982 · Cited by 97 · see citing works
Unsupervised Learning of Visual Structure using Predictive Generative NetworksPreprint
Lotter, W. E., Kreiman, G., & Cox, D. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1511.06380 · Cited by 97 · see citing works
Supervised and Unsupervised Learning with Two Sites of Synaptic Integration
Kording, K. P., & Peter König, B. (2001) · Journal of Computational Neuroscience · doi:10.1023/a:1013776130161 · Cited by 98 · see citing works
“Variants of these algorithms are based on spatial (Becker and Hinton, 1992; Stone and Bray, 1995), cross-modal (de Sa and Ballard, 1998), or temporal (Földiak, 1991) smoothness criteria and also use a second variable for each cell containing the derivative of a local objective function with respect to the cell’s activ…”
MethodLearning the invariance properties of complex cells from their responses to natural stimuli
Einhäuser, W., Kayser, C., König, P., & Körding, K. P. (2002) · European Journal of Neuroscience · doi:10.1046/j.0953-816x.2001.01885.x · Cited by 91 · see citing works
Perceptual History Influences Neural Responses to Face and Body Postures
Jellema, T., & Perrett, D. I. (2003) · Journal of Cognitive Neuroscience · doi:10.1162/089892903770007353 · Cited by 92 · see citing works
“The notion of the contiguity between different views of moving objects is encapsulated in the ‘‘temporal association hypothesis’’ (Perrett & Oram, 1993; Rolls, 1992), which has been successfully implemented in neural network models of recognition, especially those models that seek to explain transform-invariant object…”
BackgroundRepresentation of Visual Features of Objects in the Inferotemporal Cortex
Tanaka, K. (1996) · Neural Networks · doi:10.1016/s0893-6080(96)00045-7 · Cited by 94 · see citing works
Bilinear Sparse Coding for Invariant Vision
Grimes, D. B., & Rao, R. P. N. (2004) · Neural Computation · doi:10.1162/0899766052530893 · Cited by 93 · see citing works
“Examples of this approach include weight sharing and weight decay models Hinton, 1987; Földiák, 1991), the neocognitron (Fukushima, Miyake, & Takayukiito, 1983), LeNet (LeCun et al., 1989), and slow feature analysis (Wiskott & Sejnowski, 2002). A key property of the sparse bilinear model is that it preserves informati…”
MethodNonlinear ICA of temporally dependent stationary sources
Hyvärinen, A., & Morioka, H. (2017) · Työväentutkimus Vuosikirja · Cited by 90 · see citing works
- Authors
- Földiák, P.
- Affiliations
- Physiological Laboratory, University of Cambridge, Downing Street, Cambridge CB2 3EG, U.K.
- Venue
- Neural Computation
- Publisher
- MIT Press - Journals
- Published
- 1991-06
- Volume
- 3
- Issue
- 2
- Pages
- 194-200
- ISSN
- 0899-7667 (ISSN-L)
- 1530-888X
- OpenAlex
- W2096388912
- Cited by
- 661
- References
- 15
- Citations by year
- 2026: 9 · 2025: 6 · 2024: 10 · 2023: 12 · 2022: 11 · 2021: 20 · 2020: 16 · 2019: 18 · 2018: 21 · 2017: 16 · 2016: 37 · 2015: 20 · 2014: 31 · 2013: 24 · 2012: 28
- FWCI
- 5.77
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Neural Computation (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Visual perception and processing mechanisms
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Keywords
- invariance learning, transformation sequences, translation invariance, local learning rules, visual object recognition, complex cells, primary visual cortex, visual system development
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
