The desire to reduce the dependence on curated, labeled datasets and to leverage the vast quantities of unlabeled data has triggered renewed interest in unsupervised (or self-supervised) learning algorithms. Despite improved performance due to approaches such as the identification of disentangled latent representations, contrastive learning and clustering optimizations, unsupervised machine learning still falls short of its hypothesized potential as a breakthrough paradigm enabling generally intelligent systems. Inspiration from cognitive (neuro)science has been based mostly on adult learners with access to labels and a vast amount of prior knowledge. To push unsupervised machine learning forward, we argue that developmental science of infant cognition might hold the key to unlocking the next generation of unsupervised learning approaches. We identify three crucial factors enabling infants’ quality and speed of learning: (1) babies’ information processing is guided and constrained; (2) babies are learning from diverse, multimodal inputs; and (3) babies’ input is shaped by development and active learning. We assess the extent to which these insights from infant learning have already been exploited in machine learning, examine how closely these implementations resemble the core insights, and propose how further adoption of these factors can give rise to previously unseen performance levels in unsupervised learning.
This is a preview of subscription content, access via your institution
Subscribe to Nature+
Get immediate online access to Nature and 55 other Nature journal
Subscribe to Journal
Get full journal access for 1 year
only $8.25 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
Bengio, Y., Lamblin, P., Popovici, D. & Larochelle, H. Greedy layer-wise training of deep networks. In Proc. Advances in Neural Information Processing Systems Vol. 19 (eds. Schölkopf, B., Platt, J. & Hoffman, T.) 153–160 (NIPS, 2006).
Hinton, G. E., Osindero, S. & Teh, Y.-W. A fast learning algorithm for deep belief nets. Neural Comput. 18, 1527–1554 (2006).
Baldi, P. Autoencoders, unsupervised learning and deep architectures. In Proc. ICML Workshop on Unsupervised and Transfer Learning (eds. Guyon, I., Dror, G., Lemaire, V., Taylor, G. & Silver, D.) 37–49 (JMLR, 2012).
Bengio, Y., Courville, A. & Vincent, P. Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35, 1798–1828 (2013).
Erhan, D. et al. Why does unsupervised pre-training help deep learning? J. Mach. Learn. Res. 11, 625–660 (2010).
Russakovsky, O. et al. ImageNet large scale visual recognition challenge. Int. J. Comput. Vision 115, 211–252 (2015).
Carreira, J. & Zisserman, A. Quo vadis, action recognition? A new model and the kinetics dataset. In Proc. IEEE Conference on Computer Vision and Pattern Recognition 6299–6308 (IEEE, 2017).
Long, J., Shelhamer, E. & Darrell, T. Fully convolutional networks for semantic segmentation. In Proc. IEEE Conference on Computer Vision and Pattern Recognition 3431–3440 (IEEE, 2015).
Ren, S., He, K., Girshick, R. & Sun, J. Faster R-CNN: towards real-time object detection with region proposal networks. In Advances in Neural Information Processing Systems Vol. 28 (eds. Cortes, C., Lawrence, N., Lee, D., Sugiyama, M. & Garnett, R.) 91–99 (NIPS, 2015).
He, K., Girshick, R. & Dollár, P. Rethinking ImageNet pre-training. In Proc. IEEE International Conference on Computer Vision 4918–4927 (IEEE, 2019).
Huh, M., Agrawal, P. & Efros, A. A. What makes ImageNet good for transfer learning? Preprint at https://arxiv.org/abs/1608.08614 (2016).
Recht, B., Roelofs, R., Schmidt, L. & Shankar, V. Do ImageNet classifiers generalize to imagenet? In Proc. 36th International Conference on Machine Learning (eds. Chaudhuri, K. & Salakhutdinov, R.) 5389–5400 (PMLR, 2019).
Burgess, C. P. et al. Understanding disentangling in β-VAE. Preprint at https://arxiv.org/abs/1804.03599 (2018).
Caron, M., Bojanowski, P., Joulin, A. & Douze, M. Deep clustering for unsupervised learning of visual features. In Proc. European Conference on Computer Vision (eds. Ferrari, V., Hebert, M. I., Sminchisescu, C. & Weiss, Y.) 132–149 (Springer, 2018).
Chen, T., Kornblith, S., Norouzi, M. & Hinton, G. A simple framework for contrastive learning of visual representations. In Proc. 37th International Conference on Machine Learning (eds. Daumé, H. III & Singh, A.) 1597–1607 (PMLR, 2020).
Zbontar, J., Jing, L., Misra, I., LeCun, Y. & Deny, S. Barlow twins: self-supervised learning via redundancy reduction. In Proc. 38th International Conference on Machine Learning (eds. Meila, M. & Zhang, T.) 12310–12320 (PMLR, 2021).
Ma, W. J. & Peters, B. A neural network walks into a lab: towards using deep nets as models for human behavior. Preprint at https://arxiv.org/abs/2005.02181 (2020).
Yamins, D. L. & DiCarlo, J. J. Using goal-driven deep learning models to understand sensory cortex. Nat. Neurosci. 19, 356–365 (2016).
Macpherson, T. et al. Natural and artificial intelligence: a brief introduction to the interplay between AI and neuroscience research. Neural Networks 144, 603–613 (2021).
Rosenblatt, F. The perceptron: a probabilistic model for information storage and organization in the brain. Psychol. Rev. 65, 386–408 (1958).
McCulloch, W. S. & Pitts, W. A logical calculus of the ideas immanent in nervous activity. Bull. Math. Biophys. 5, 115–133 (1943).
Hassabis, D., Kumaran, D., Summerfield, C. & Botvinick, M. Neuroscience-inspired artificial intelligence. Neuron 95, 245–258 (2017).
Lake, B. M., Ullman, T. D., Tenenbaum, J. B. & Gershman, S. J. Building machines that learn and think like people. Behav. Brain Sci. 40, e253 (2017).
Saxe, A., Nelli, S. & Summerfield, C. If deep learning is the answer, then what is the question? Nat. Rev. Neurosci. 22, 55–67 (2021).
Sinz, F. H., Pitkow, X., Reimer, J., Bethge, M. & Tolias, A. S. Engineering a less artificial intelligence. Neuron 103, 967–979 (2019).
Zador, A. M. A critique of pure learning and what artificial neural networks can learn from animal brains. Nat. Commun. 10, 3770 (2019).
Bengio, Y. Deep learning of representations for unsupervised and transfer learning. In Proc. ICML Workshop on Unsupervised and Transfer Learning (eds. Guyon, I., Dror, G., Lemaire, V., Taylor, G. & Silver, D.) 17–36 (JMLR, 2012).
Cangelosi, A. & Schlesinger, M. Developmental Robotics: From Babies to Robots (MIT Press, 2015).
Kidd, C. How to know. In Proc. 33rd Conference on Neural Information Processing Systems (NIPS, 2019).
Gopnik, A. An AI that knows the world like children do. Sci. Am. Mind 28, 21–28 (2017).
Kosoy, E. et al. Exploring exploration: comparing children with RL agents in unified environments. Preprint at https://arxiv.org/abs/2005.02880 (2020).
Smith, L. B. & Breazeal, C. The dynamic lift of developmental process. Dev. Sci. 10, 61–68 (2007).
Smith, L. B. & Slone, L. K. A developmental approach to machine learning? Front. Psychol. 8, 2124 (2017).
Tenenbaum, J. B., Kemp, C., Griffiths, T. L. & Goodman, N. D. How to grow a mind: statistics, structure and abstraction. Science 331, 1279–1285 (2011).
Adolph, K. E., Hoch, J. E. & Cole, W. G. Development (of walking): 15 suggestions. Trends Cogn. Sci. 22, 699–711 (2018).
Byrge, L., Sporns, O. & Smith, L. B. Developmental process emerges from extended brain-body-behavior networks. Trends Cogn. Sci. 18, 395–403 (2014).
Hunnius, S. Early cognitive development: five lessons from infant learning. In Oxford Research Encyclopedia of Psychology (ed. Braddick, O.) (Oxford Univ. Press, in the press).
Karmiloff-Smith, A. An alternative to domain-general or domain-specific frameworks for theorizing about human evolution and ontogenesis. AIMS Neurosci. 2, 91–104 (2015).
von Hofsten, C. & Rosander, K. The development of sensorimotor intelligence in infants. Adv. Child Dev. Behav. 55, 73–106 (2018).
Dunsworth, H. M., Warrener, A. G., Deacon, T., Ellison, P. T. & Pontzer, H. Metabolic hypothesis for human altriciality. Proc. Natl Acad. Sci. USA 109, 15212–15216 (2012).
Haeusler, M. et al. The obstetrical dilemma hypothesis: there’s life in the old dog yet. Biol. Rev. 96, 2031–2057 (2021).
Bethlehem, R. A. et al. Brain charts for the human lifespan. Nature 604, 525–533 (2022).
Huttenlocher, P. R. et al. Synaptic density in human frontal cortex-developmental changes and effects of aging. Brain Res. 163, 195–205 (1979).
Deoni, S. C. et al. Mapping infant brain myelination with magnetic resonance imaging. J. Neurosci. 31, 784–791 (2011).
Hill, J. et al. A surface-based analysis of hemispheric asymmetries and folding of cerebral cortex in term-born human infants. J. Neurosci. 30, 2268–2276 (2010).
Clouchoux, C. et al. Quantitative in vivo MRI measurement of cortical development in the fetus. Brain Struct. Funct. 217, 127–139 (2012).
Cabral, L., Zubiaurre, L., Wild, C., Linke, A. & Cusack, R. Category-selective visual regions have distinctive signatures of connectivity in early infancy. Prerpint at https://www.biorxiv.org/content/10.1101/675421v2.full (2019).
Doria, V. et al. Emergence of resting state networks in the preterm human brain. Proc. Natl Acad. Sci. USA 107, 20015–20020 (2010).
Kamps, F. S., Hendrix, C. L., Brennan, P. A. & Dilks, D. D. Connectivity at the origins of domain specificity in the cortical face and place networks. Proc. Natl Acad. Sci. USA 117, 6163–6169 (2020).
Cusack, R., Wild, C. J., Zubiaurre-Elorza, L. & Linke, A. C. Why does language not emerge until the second year? Hearing Res. 366, 75–81 (2018).
Deen, B. et al. Organization of high-level visual cortex in human infants. Nat. Commun. 8, 13995 (2017).
Ellis, C. T. et al. Evidence of hippocampal learning in human infants. Curr. Biol. 31, 3358–3364 (2021).
Ellis, C. T., Skalaban, L. J., Yates, T. S. & Turk-Browne, N. B. Attention recruits frontal cortex in human infants. Proc. Natl Acad. Sci. USA 118, e2021474118 (2021).
Raz, G. & Saxe, R. Learning in infancy is active, endogenously motivated, and depends on the prefrontal cortices. Annu. Rev. Dev. Psychol. 2, 247–268 (2020).
Linke, A. C. et al. Disruption to functional networks in neonates with perinatal brain injury predicts motor skills at 8 months. NeuroImage Clin. 18, 399–406 (2018).
Marcus, G. F., Vijayan, S., Rao, S. B. & Vishton, P. M. Rule learning by seven-month-old infants. Science 283, 77–80 (1999).
Elman, J. L. Finding structure in time. Cogn. Sci. 14, 179–211 (1990).
Alhama, R. G. & Zuidema, W. Pre-wiring and pre-training: what does a neural network need to learn truly general identity rules? J. Artif. Intell. Res. 61, 927–946 (2018).
Jeffress, L. A. A place theory of sound localization. J. Comp. Physiol. Psychol. 41, 35–39 (1948).
Jaeger, H. The ‘Echo State’ Approach to Analysing and Training Recurrent Neural Networks—With an Erratum Note. Technical Report 148, 13 (German National Research Center for Information Technology (GMD), 2001).
Smith, L. B. Do infants possess innate knowledge structures? The con side. Dev. Sci. 2, 133–144 (1999).
Spelke, E. Initial knowledge: six suggestions. Cognition 50, 431–445 (1995).
Stahl, A. E. & Feigenson, L. Observing the unexpected enhances infants’ learning and exploration. Science 348, 91–94 (2015).
Simion, F., Di Giorgio, E., Leo, I. & Bardi, L. The processing of social stimuli in early infancy: from faces to biological motion perception. In Progress in Brain Research Vol. 189 (eds. Braddick, O., Atkinson, J. & Innocenti, G. M.) 173–193 (Elsevier, 2011).
Reynolds, G. D. & Roth, K. C. The development of attentional biases for faces in infancy: a developmental systems perspective. Front. Psychol. 9, 222 (2018).
Viola Macchi, C., Turati, C. & Simion, F. Can a nonspecific bias toward top-heavy patterns explain newborns’ face preference? Psychol. Sci. 15, 379–383 (2004).
Chien, S. H.-L. No more top-heavy bias: Infants and adults prefer upright faces but not top-heavy geometric or face-like patterns. J. Vision 11, 13 (2011).
Ichikawa, H., Tsuruhara, A., Kanazawa, S. & Yamaguchi, M. K. Two- to three-month-old infants prefer moving face patterns to moving top-heavy patterns. Jap. Psychol. Res. 55, 254–263 (2013).
Cooper, R. P. & Aslin, R. N. Preference for infant-directed speech in the first month after birth. Child Dev. 61, 1584–1595 (1990).
Peña, M. et al. Sounds and silence: an optical topography study of language recognition at birth. Proc. Natl Acad. Sci. USA 100, 11702–11705 (2003).
Vouloumanos, A. & Werker, J. F. Listening to language at birth: evidence for a bias for speech in neonates. Dev. Sci. 10, 159–164 (2007).
Mély, D. A., Linsley, D. & Serre, T. Complementary surrounds explain diverse contextual phenomena across visual modalities. Psychol. Rev. 125, 769 (2018).
Linsley, D., Kim, J., Ashok, A. & Serre, T. Recurrent neural circuits for contour detection. In Proc. 8th International Conference on Learning Representations (ICLR, 2020).
Michalski, R. S. in Machine Learning 83–134 (Morgan Kaufmann, 1983).
Mitchell, T. The Need for Biases in Learning Generalizations. Rutgers Computer Science Technical Report cbm-tr-117 (Rutgers University, 1980).
Feinman, R. & Lake, B. M. Learning inductive biases with simple neural networks. In Proc. 40th Annual Meeting of the Cognitive Science Society (eds. Kalish, C., Rau, M. A., Zhu, X. & Rogers, T. T.) (CSS, 2018).
Kopparti, R. M. & Weyde, T. Weight priors for learning identity relations. In Advances in Neural Information Processing Systems Vol. 33 (eds. Wallach, H. M., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F. E. Fox, A. & Garnett, R.) (NIPS, 2020).
Weyde, T. & Kopparti, R. M. Modelling identity rules with neural networks. J. Appl. Logics 6, 745–769 (2019).
Ullman, S., Harari, D. & Dorfman, N. From simple innate biases to complex visual concepts. Proc. Natl Acad. Sci. USA 109, 18215–18220 (2012).
Felleman, D. J. & Van Essen, D. C. Distributed hierarchical processing in the primate cerebral cortex. Cereb. Cortex 1, 1–47 (1991).
Krizhevsky, A., Sutskever, I. & Hinton, G. E. ImageNet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst. 25, 1097–1105 (2012).
He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. In Proc. IEEE Conference on Computer Vision and Pattern Recognition 770–778 (IEEE, 2016).
Szegedy, C. et al. Going deeper with convolutions. In Proc. IEEE Conference on Computer Vision and Pattern Recognition 1–9 (IEEE, 2015).
Besold, T. R. et al. Neural-symbolic learning and reasoning: a survey and interpretation. In Neuro-Symbolic Artificial Intelligence: The State of the Art (eds. Hitzler, P. & Sarker, M. K.) 1–51 (IOS Press, 2021).
d’Avila Garcez, A. S. & Gabbay, D. M. Fibring neural networks. In Proc. Nineteenth National Conference on Artificial Intelligence, Sixteenth Conference on Innovative Applications of Artificial Intelligence (eds. McGuinness, D. L. & Ferguson, G.) 342–347 (AAAI Press/MIT Press, 2004).
Saffran, J. R. & Kirkham, N. Z. Infant statistical learning. Annu. Rev. Psychol. 69, 181–203 (2018).
Teinonen, T., Fellman, V., Näätänen, R., Alku, P. & Huotilainen, M. Statistical language learning in neonates revealed by event-related brain potentials. BMC Neurosci. 10, 21 (2009).
Jacquey, L., Fagard, J., Esseily, R. & O’Regan, J. K. Detection of sensorimotor contingencies in infants before the age of one year: a comprehensive review. Dev. Psychol 56, 1233–1251 (2020).
Zaadnoordijk, L. et al. From movement to action: an EEG study into the emerging sense of agency in early infancy. Dev. Cogn. Neurosci. 42, 100760 (2020).
Hunnius, S. & Bekkering, H. The early development of object knowledge: a study of infants’ visual anticipations during action observation. Dev. Psychol. 46, 446–454 (2010).
Brookes, H. et al. Three-month-old infants learn arbitrary auditory-visual pairings between voices and faces. Infant Child Dev. 10, 75–82 (2001).
Gómez, R. & Maye, J. The developmental trajectory of nonadjacent dependency learning. Infancy 7, 183–206 (2005).
Maye, J., Werker, J. F. & Gerken, L. Infant sensitivity to distributional information can affect phonetic discrimination. Cognition 82, B101–B111 (2002).
Saffran, J. R., Aslin, R. N. & Newport, E. L. Statistical learning by 8-month-old infants. Science 274, 1926–1928 (1996).
Emberson, L. L., Misyak, J. B., Schwade, J. A., Christiansen, M. H. & Goldstein, M. H. Comparing statistical learning across perceptual modalities in infancy: an investigation of underlying learning mechanism(s). Dev. Sci. 22, e12847 (2019).
Kirkham, N. Z., Slemmer, J. A. & Johnson, S. P. Visual statistical learning in infancy: evidence for a domain general learning mechanism. Cognition 83, B35–B42 (2002).
Monroy, C. D. et al. Sensitivity to structure in action sequences: an infant event-related potential study. Neuropsychologia 126, 92–101 (2019).
Stahl, A. E., Romberg, A. R., Roseberry, S., Golinkoff, R. M. & Hirsh-Pasek, K. Infants segment continuous events using transitional probabilities. Child Dev. 85, 1821–1826 (2014).
Tummeltshammer, K. S. & Kirkham, N. Z. Learning to look: probabilistic variation and noise guide infants’ eye movements. Dev. Sci. 16, 760–771 (2013).
Ruffman, T., Taumoepeau, M. & Perkins, C. Statistical learning as a basis for social understanding in children. Br. J. Dev. Psychol. 30, 87–104 (2012).
Bristow, D. et al. Hearing faces: how the infant brain matches the face it sees with the speech it hears. J. Cogn. Neurosci. 21, 905–921 (2008).
Bremner, A. J., Mareschal, D., Lloyd-Fox, S. & Spence, C. Spatial localization of touch in the first year of life: early influence of a visual spatial code and the development of remapping across changes in limb position. J. Exp. Psychol. Gen. 137, 149–162 (2008).
Zmyj, N., Jank, J., Schütz-Bosbach, S. & Daum, M. M. Detection of visual-tactile contingency in the first year after birth. Cognition 120, 82–89 (2011).
Tanaka, Y., Kanakogi, Y., Kawasaki, M. & Myowa, M. The integration of audio-tactile information is modulated by multimodal social interaction with physical contact in infancy. Dev. Cogn. Neurosci. 30, 31–40 (2018).
Lewkowicz, D. J. The development of intersensory temporal perception: an epigenetic systems/limitations view. Psychol. Bull. 126, 281–308 (2000).
Landry, S. P., Guillemot, J.-P. & Champoux, F. Temporary deafness can impair multisensory integration: a study of cochlear-implant users. Psychol. Sci. 24, 1260–1268 (2013).
Stevenson, R., Sheffield, S. W., Butera, I. M., Gifford, R. H. & Wallace, M. Multisensory integration in cochlear implant recipients. Ear Hearing 38, 521–538 (2017).
Weatherhead, D. & White, K. S. Read my lips: visual speech influences word processing in infants. Cognition 160, 103–109 (2017).
Cappagli, G., Cocchi, E. & Gori, M. Auditory and proprioceptive spatial impairments in blind children and adults. Dev. Sci. 20, e12374 (2017).
Bruni, E., Tran, N.-K. & Baroni, M. Multimodal distributional semantics. J. Artif. Intell. Res. 49, 1–47 (2014).
Marton, Z.-C., Pangercic, D., Blodow, N. & Beetz, M. Combined 2D-3D categorization and classification for multimodal perception systems. Int. J. Robot. Res. 30, 1378–1402 (2011).
Nakamura, T., Nagai, T. & Iwahashi, N. Multimodal object categorization by a robot. In Proc. 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems 2415–2420 (IEEE, 2007).
Mogadala, A., Kalimuthu, M. & Klakow, D. Trends in integration of vision and language research: a survey of tasks, datasets, and methods. J. Artif. Intell. Res. 71, 1183–1317 (2021).
Barbieri, F. et al. Towards a multimodal time-based empathy prediction system. In Proc. 2019 14th IEEE International Conference on Automatic Face and Gesture Recognition 1–5 (IEEE, 2019).
Tzirakis, P., Trigeorgis, G., Nicolaou, M. A., Schuller, B. W. & Zafeiriou, S. End-to-end multimodal emotion recognition using deep neural networks. IEEE J. Select. Top. Signal Process. 11, 1301–1309 (2017).
Evangelopoulos, G. et al. Multimodal saliency and fusion for movie summarization based on aural, visual and textual attention. IEEE Trans. Multimedia 15, 1553–1568 (2013).
de Sa, V. R. & Ballard, D. H. Category learning through multimodality sensing. Neural Comput. 10, 1097–1117 (1998).
Radford, A., Metz, L. & Chintala, S. Unsupervised representation learning with deep convolutional generative adversarial networks. In Proc. 4th International Conference on Learning Representations (eds. Bengio, Y. & LeCun, Y.) (ICLR, 2016).
Droniou, A., Ivaldi, S. & Sigaud, O. Deep unsupervised network for multimodal perception, representation and classification. Robot. Auton. Syst. 71, 83–98 (2015).
Feng, Y., Ma, L., Liu, W. & Luo, J. Unsupervised image captioning. In Proc. IEEE Conference on Computer Vision and Pattern Recognition 4125–4134 (IEEE, 2019).
Ngiam, J. et al. Multimodal deep learning. In Proc. 28th International Conference on International Conference on Machine Learning (eds. Getoor, L. & Scheffer, T.) 689–696 (Omnipress, 2011).
Srivastava, N. & Salakhutdinov, R. R. Multimodal learning with deep Boltzmann machines. In Advances in Neural Information Processing Systems (Pereira, F., Burges, C. J., Bottou, L. & Weinberger, K. Q.) 2222–2230 (NIPS, 2012).
Bachman, P., Hjelm, R. D. & Buchwalter, W. Learning representations by maximizing mutual information across views. In Advances in Neural Information Processing Systems (eds. Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E. & Garnett, R.) 15509–15519 (NIPS, 2019).
Tian, Y., Krishnan, D. & Isola, P. Contrastive multiview coding. In Proc. Computer Vision–ECCV 2020: 16th European Conference Part XI 16 (Vedaldi, A., Bischof, H., Brox, T. & Frahm, J.-M.) 776–794 (Springer, 2020).
Roads, B. D. & Love, B. C. Learning as the unsupervised alignment of conceptual systems. Nat. Mach. Intell. 2, 76–82 (2020).
Wang, C. & Mahadevan, S. Manifold alignment without correspondence. In Proc. 21st International Jont Conference on Artifical Intelligence (ed. Boutilier, C.) 1273–1278 (Morgan Kaufmann, 2009).
Wang, C. & Mahadevan, S. Heterogeneous domain adaptation using manifold alignment. In Proc. Twenty-Second International Joint Conference on Artificial Intelligence Vol. 2 (ed. Walsh, T.) 1541–1546 (AAAI Press, 2011).
Baltrušaitis, T., Ahuja, C. & Morency, L.-P. Multimodal machine learning: a survey and taxonomy. IEEE Trans. Pattern Anal. Mach. Intell. 41, 423–443 (2018).
Barros, P., Eppe, M., Parisi, G. I., Liu, X. & Wermter, S. Expectation learning for stimulus prediction across modalities improves unisensory classification. Front. Robot. AI 6, 137 (2019).
Peterson, S. M., Rao, R. P. & Brunton, B. W. Learning neural decoders without labels using multiple data streams. Preprint at https://www.biorxiv.org/content/10.1101/2021.09.10.459775v1.full (2021).
Ackman, J. B., Burbridge, T. J. & Crair, M. C. Retinal waves coordinate patterned activity throughout the developing visual system. Nature 490, 219–225 (2012).
Moon, C., Lagercrantz, H. & Kuhl, P. K. Language experienced in utero affects vowel perception after birth: a two-country study. Acta Paediatrica 102, 156–160 (2013).
DeCasper, A. J. & Spence, M. J. Prenatal maternal speech influences newborns’ perception of speech sounds. Infant Behav. Dev. 9, 133–150 (1986).
Lobo, M. A., Kokkoni, E., de Campos, A. C. & Galloway, J. C. Not just playing around: infants’ behaviors with objects reflect ability, constraints and object properties. Infant Behav. Dev. 37, 334–351 (2014).
Soska, K. C. & Adolph, K. E. Postural position constrains multimodal object exploration in infants. Infancy 19, 138–161 (2014).
Campos, J. J. et al. Travel broadens the mind. Infancy 1, 149–219 (2000).
Barsalou, L. W. Grounded cognition. Annu. Rev. Psychol. 59, 617–645 (2008).
Dobson, V. & Teller, D. Y. Visual acuity in human infants: a review and comparison of behavioral and electrophysiological studies. Vision Res. 18, 1469–1483 (1978).
Sokol, S. Measurement of infant visual acuity from pattern reversal evoked potentials. Vision Res. 18, 33–39 (1978).
Fiser, J., Aslin, R., Lathrop, A., Rothkopf, C. & Markant, J. An infants’ eye view of the world: implications for learning in natural contexts. In Proc. International Conference on Infant Studies (2006).
Franchak, J. M., Kretch, K. S., Soska, K. C. & Adolph, K. E. Head-mounted eye tracking: a new method to describe infant looking. Child Dev. 82, 1738–1750 (2011).
Smith, L. B., Yu, C., Yoshida, H. & Fausey, C. M. Contributions of head-mounted cameras to studying the visual environments of infants and young children. J. Cogn. Dev. 16, 407–419 (2015).
Yoshida, H. & Smith, L. B. What’s in view for toddlers? Using a head camera to study visual experience. Infancy 13, 229–248 (2008).
Smith, L. B., Jayaraman, S., Clerkin, E. & Yu, C. The developing infant creates a curriculum for statistical learning. Trends Cogn. Sci. 22, 325–336 (2018).
Fausey, C. M., Jayaraman, S. & Smith, L. B. From faces to hands: changing visual input in the first two years. Cognition 152, 101–107 (2016).
Davis, J. et al. Does neonatal imitation exist? Insights from a meta-analysis of 336 effect sizes. Perspect. Psychol. Sci. https://doi.org/10.1177/1745691620959834 (2021).
Hunnius, S. & Bekkering, H. What are you doing? How active and observational experience shape infants’ action understanding. Philos. Trans. R. Soc. B Biol. Sci. 369, 20130490 (2014).
Meltzoff, A. N. & Moore, M. K. Explaining facial imitation: a theoretical model. Infant Child Dev. 6, 179–192 (1997).
Meltzoff, A. N. & Marshall, P. J. Human infant imitation as a social survival circuit. Curr. Opin. Behav. Sci. 24, 130–136 (2018).
Ray, E. & Heyes, C. Imitation in infancy: the wealth of the stimulus. Dev. Sci. 14, 92–105 (2011).
Soderstrom, M. Beyond babytalk: re-evaluating the nature and content of speech input to preverbal infants. Dev. Rev. 27, 501–532 (2007).
Brand, R. J., Baldwin, D. A. & Ashburn, L. A. Evidence for ‘motionese’: modifications in mothers’ infant-directed action. Dev. Sci. 5, 72–83 (2002).
van Schaik, J. E., Meyer, M., van Ham, C. R. & Hunnius, S. Motion tracking of parents’ infant-versus adult-directed actions reveals general and action-specific modulations. Dev. Sci. 23, e12869 (2020).
Wass, S. V. et al. Infants’ visual sustained attention is higher during joint play than solo play: is this due to increased endogenous attention control or exogenous stimulus capture? Dev. Sci. 21, e12667 (2018).
Yu, C. & Smith, L. B. The social origins of sustained attention in one-year-old human infants. Curr. Biol. 26, 1235–1240 (2016).
Yu, Y. et al. The theoretical and methodological opportunities afforded by guided play with young children. Front. Psychol. 9, 1152 (2018).
Bazhydai, M., Westermann, G. & Parise, E. ‘I don’t know but I know who to ask’: 12-month-olds actively seek information from knowledgeable adults. Dev. Sci. 23, e12938 (2020).
Poulin-Dubois, D. & Brosseau-Liard, P. The developmental origins of selective social learning. Curr. Directions Psychol. Sci. 25, 60–64 (2016).
Berlyne, D. E. Conflict, Arousal and Curiosity (McGraw-Hill, 1960).
Day, H. I. Curiosity and the interested explorer. Performance & Instruction 21, 19–22 (1982).
Kidd, C., Piantadosi, S. T. & Aslin, R. N. The Goldilocks effect: human infants allocate attention to visual sequences that are neither too simple nor too complex. PLoS ONE 7, e36399 (2012).
Kidd, C., Piantadosi, S. T. & Aslin, R. N. The Goldilocks effect in infant auditory attention. Child Dev. 85, 1795–1804 (2014).
Poli, F., Serino, G., Mars, R. & Hunnius, S. Infants tailor their attention to maximize learning. Sci. Adv. 6, eabb5053 (2020).
Cohen, L. B. Uses and misuses of habituation and related preference paradigms. Infant Child Dev. 13, 349–352 (2004).
Hunter, M. A. & Ames, E. W. A multifactor model of infant preferences for novel and familiar stimuli. Adv. Infancy Res 5, 69–95 (1988).
Aslin, R. N. What’s in a look? Dev. Sci. 10, 48–53 (2007).
Haith, M. M. Who put the cog in infant cognition? Is rich interpretation too costly? Infant Behav. Dev. 21, 167–179 (1998).
Adolph, K. E. et al. How do you learn to walk? Thousands of steps and dozens of falls per day. Psychol. Sci. 23, 1387–1394 (2012).
Hoch, J. E., O’Grady, S. M. & Adolph, K. E. It’s the journey, not the destination: locomotor exploration in infants. Dev. Sci. 22, e12740 (2019).
Oakes, L. M. & Baumgartner, H. A. Manual object exploration and learning about object features in human infants. In Proc. 2012 IEEE International Conference on Development and Learning and Epigenetic Robotics1–6 (IEEE, 2012).
Elman, J. L. Learning and development in neural networks: the importance of starting small. Cognition 48, 71–99 (1993).
Bengio, Y., Louradour, J., Collobert, R. & Weston, J. Curriculum learning. In Proc. 26th Annual International Conference on Machine Learning (eds. Pohoreckyj, A., Danyluk, L. Bottou, M. & Littman, L.) 41–48 (ACM, 2009).
Vogelsang, L. et al. Potential downside of high initial visual acuity. Proc. Natl Acad. Sci. USA 115, 11333–11338 (2018).
Orhan, A. E., Gupta, V. V. & Lake, B. M. Self-supervised learning through the eyes of a child. In Advances in Neural Information Processing Systems Vol. 33 (eds. Wallach, H. M., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E. A. & Garnett, R.) 9960–9971 (NIPS, 2020).
Newport, E. L., Bavelier, D. & Neville, H. J. Critical thinking about critical periods: perspectives on a critical period for language acquisition. In Language, Brain and Cognitive Development: Essays in Honor of Jacques Mehler (ed. Dupoux, E.) 481–502 (MIT Press, 2001).
Molnár, Z., Luhmann, H. J. & Kanold, P. O. Transient cortical circuits match spontaneous and sensory-driven activity during development. Science 370, eabb2153 (2020).
Kostovic, I. & Rakic, P. Developmental history of the transient subplate zone in the visual and somatosensory cortex of the macaque monkey and human brain. J. Comp. Neurol. 297, 441–470 (1990).
Achille, A., Rovere, M. & Soatto, S. Critical learning periods in deep neural networks. In Proc. 7th International Conference on Learning Representations (ICLR, 2019).
Carpenter, G. A. & Grossberg, S. The art of adaptive pattern recognition by a self-organizing neural network. Computer 21, 77–88 (1988).
French, R. M. Catastrophic forgetting in connectionist networks. Trends Cogn. Sci. 3, 128–135 (1999).
Parisi, G. I., Kemker, R., Part, J. L., Kanan, C. & Wermter, S. Continual lifelong learning with neural networks: a review. Neural Networks 113, 54–71 (2019).
Robins, A. Catastrophic forgetting, rehearsal and pseudorehearsal. Connection Sci. 7, 123–146 (1995).
Hinton, G. E. & Plaut, D. C. Using fast weights to deblur old memories. In Proc. 9th Annual Conference of the Cognitive Science Society 177–186 (Erlbaum, 1987).
Kemker, R. & Kanan, C. FearNet: brain-inspired model for incremental learning. In Proc. 6th International Conference on Learning Representations (ICLR, 2018).
Kirkpatrick, J. et al. Overcoming catastrophic forgetting in neural networks. Proc. Natl Acad. Sci. USA 114, 3521–3526 (2017).
Rannen, A., Aljundi, R., Blaschko, M. B. & Tuytelaars, T. Encoder based lifelong learning. In Proc. IEEE International Conference on Computer Vision 1320–1328 (IEEE, 2017).
Draelos, T. J. et al. Neurogenesis deep learning: extending deep networks to accommodate new classes. In Proc. 2017 International Joint Conference on Neural Networks 526–533 (IEEE, 2017).
Javed, K. & White, M. Meta-learning representations for continual learning. In Advances in Neural Information Processing Systems (eds. Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E. & Garnett, R.) 1818–1828 (NIPS, 2019).
Kemker, R., McClure, M., Abitino, A., Hayes, T. L. & Kanan, C. Measuring catastrophic forgetting in neural networks. In Proc. Thirty-Second AAAI Conference on Artificial Intelligence Vol. 415 (eds. McIlraith, S. A. & Weinberger, K. Q.) 3390–3398 (AAAI, 2018).
Settles, B. Active Learning Literature Survey (Univ. Wisconsin-Madison Department of Computer Sciences, 2009).
Settles, B. From theories to queries: active learning in practice. In Proc. Active Learning and Experimental Design Workshop in Conjunction with AISTATS 2010 Vol. 16 (eds. Guyon, I., Cawley, G., Dror, G., Lemaire, V. & Statnikov, A.) 1–18 (MLR, 2011).
Botvinick, M. M., Niv, Y. & Barto, A. G. Hierarchically organized behavior and its neural foundations: a reinforcement learning perspective. Cognition 113, 262–280 (2009).
Lefort, M. & Gepperth, A. Active learning of local predictable representations with artificial curiosity. In Proc. 2015 Joint IEEE International Conference on Development and Learning and Epigenetic Robotics 228–233 (IEEE, 2015).
Graves, A., Bellemare, M. G., Menick, J., Munos, R. & Kavukcuoglu, K. Automated curriculum learning for neural networks. In Proc. 34th International Conference on Machine Learning Vol. 70 (eds. Precup, D. & Teh, Y. W.) 1311–1320 (JMLR, 2017).
Schmidhuber, J. Driven by compression progress: a simple principle explains essential aspects of subjective beauty, novelty, surprise, interestingness, attention, curiosity, creativity, art, science, music, jokes. In Proc. Workshop on Anticipatory Behavior in Adaptive Learning Systems (eds. Pezzulo, G. Butz, M. V., Sigaud, O. & Baldassarre, G.) 48–76 (Springer, 2008).
Oudeyer, P.-Y. Computational theories of curiosity-driven learning. In The New Science of Curiosity (ed. Gordon, G.) 43–72 (Nova Science, 2018).
Oudeyer, P.-Y., Kaplan, F. & Hafner, V. V. Intrinsic motivation systems for autonomous mental development. IEEE Trans. Evolution. Comput. 11, 265–286 (2007).
Twomey, K. E. & Westermann, G. Curiosity-based learning in infants: a neurocomputational approach. Dev. Sci. 21, e12629 (2018).
Haber, N., Mrowca, D., Fei-Fei, L. & Yamins, D. L. Emergence of structured behaviors from curiosity-based intrinsic motivation. Preprint at https://arxiv.org/abs/1802.07461 (2018).
Soltoggio, A., Stanley, K. O. & Risi, S. Born to learn: the inspiration, progress, and future of evolved plastic artificial neural networks. Neural Networks 108, 48–67 (2018).
Arandjelovic, R. & Zisserman, A. Look, listen and learn. In Proc. IEEE International Conference on Computer Vision 609–617 (IEEE, 2017).
Barros, P., Parisi, G. I., Weber, C. & Wermter, S. Emotion-modulated attention improves expression recognition: a deep learning model. Neurocomputing 253, 104–114 (2017).
Senocak, A., Oh, T.-H., Kim, J., Yang, M.-H. & Kweon, I. S. Learning to localize sound source in visual scenes. In Proc. IEEE Conference on Computer Vision and Pattern Recognition 4358–4366 (IEEE, 2018).
Ashby, F. G. & Vucovich, L. E. The role of feedback contingency in perceptual category learning. J. Exp. Psychol. Learn. Mem. Cogn. 42, 1731–1746 (2016).
Parisi, G. I., Tani, J., Weber, C. & Wermter, S. Lifelong learning of human actions with deep neural network self-organization. Neural Networks 96, 137–149 (2017).
Bonawitz, E. & Shafto, P. Computational models of development, social influences. Curr. Opin. Behav. Sci. 7, 95–100 (2016).
Brockman, G. et al. OpenAI gym. Preprint at https://arxiv.org/abs/1606.01540 (2016).
Geirhos, R. et al. ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness. In Proc. 7th International Conference on Learning Representations (2019).
This work was supported by an ERC Advanced Grant (FOUNDCOG, #787981) awarded to R.C. and an MSCA Individual Fellowship (InterPlay, #891535) awarded to L.Z. We thank K. Storrs and K. Körding as well as the Nature Machine Intelligence editorial team for their helpful feedback.
The authors declare no competing interests.
Peer review information
Nature Machine Intelligence thanks the anonymous reviewers for their contribution to the peer review of this work.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Zaadnoordijk, L., Besold, T.R. & Cusack, R. Lessons from infant learning for unsupervised machine learning. Nat Mach Intell 4, 510–520 (2022). https://doi.org/10.1038/s42256-022-00488-2