How to Represent Part-Whole Hierarchies in a Neural Network
Open accessThe DOI matches its Crossref record; 24 of the 24 references it cites were found in a registry or an index.
Abstract
This article does not describe a working system. Instead, it presents a single idea about representation that allows advances made by several different groups to be combined into an imaginary system called GLOM.1 The advances include transformers, neural fields, contrastive representation learning, distillation, and capsules. GLOM answers the question: How can a neural network with a fixed architecture parse an image into a part-whole hierarchy that has a different structure for each image? The idea is simply to use islands of identical vectors to represent the nodes in the parse tree. If GLOM can be made to work, it should significantly improve the interpretability of the representations produced by transformer-like systems when applied to vision or language.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472
Stochastic relaxation, gibbs distributions, and the bayesian restoration of images.
Geman, S., & Geman, D. (1984) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.1984.4767596 · Cited by 17,949 · cited 1×
“Pioneering work on using Markov Random Fields for image segmentation [Geman and Geman, 1984] used the presence of a boundary between pixel x and pixel y to prevent x from attending to y .”
MethodDistilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204
Self-citationRobust Real-Time Face Detection
Viola, P., & Jones, M. J. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000013087.49260.fb · Cited by 14,097
Momentum Contrast for Unsupervised Visual Representation Learning
He, K., Fan, H., Wu, Y., Xie, S., & Girshick, R. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00975 · Cited by 13,347
Representation Learning with Contrastive Predictive Coding
Oord, A. V. D., Li, Y., & Vinyals, O. (2018) · arXiv · doi:10.48550/arxiv.1807.03748 · Cited by 6,242
Rao, R. P. N., & Ballard, D. H. (1999) · Nature Neuroscience · doi:10.1038/4580 · Cited by 5,483
Training Products of Experts by Minimizing Contrastive Divergence
Hinton, G. E. (2002) · Neural Computation · doi:10.1162/089976602760128018 · Cited by 4,879 · cited 2×
“A much more flexible way to tune the neurons would be to use a mixture of a Gaussian and a uniform [Hinton, 2002].”
MethodSelf-citationBootstrap Your Own Latent: A New Approach to Self-Supervised Learning
Grill, J.-B., Strub, F., Altché, F., Tallec, C., Richemond, P. H., Buchatskaya, E., Doersch, C., Pires, B. A., Guo, Z. D., Azar, M. G., et al. (2020) · arXiv · doi:10.48550/arxiv.2006.07733 · Cited by 3,846 · cited 1×
“This should greatly reduce the tendency towards collapse and when combined with the deep denoising autoencoder objective function and other recent tricks [Grill et al., 2020, Chen and He, 2020] it may eliminate the need for negative examples.”
BackgroundExploring Simple Siamese Representation Learning
Chen, X., & He, K. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.01549 · Cited by 3,725
A View of the Em Algorithm that Justifies Incremental, Sparse, and other Variants
Neal, R. M., & Hinton, G. E. (1998) · Learning in Graphical Models · doi:10.1007/978-94-011-5014-9_12 · Cited by 2,187
Self-citationBuciluǎ, C., Caruana, R., & Niculescu-Mizil, A. (2006) · Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/1150402.1150464 · Cited by 2,068 · cited 1×
“We can achieve this using knowledge distillation [Buciluˇa et al., 2006, Hinton et al., 2014].”
MethodHinton, G. E., Krizhevsky, A., & Wang, S. D. (2011) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-21735-7_6 · Cited by 1,360 · cited 1×
“The transforming autoencoder [Hinton et al., 2011] forced an encoder to extract an explicit representation of pose in each capsule by insisting that the reconstructed image should be a transformed version of the original image and specifying this transformation as a matrix that multiplied whatever…”
BackgroundSelf-citationMemory of sequential experience in the hippocampus during slow wave sleep
Lee, A. K., & Wilson, M. A. (2002) · Neuron · doi:10.1016/s0896-6273(02)01096-6 · Cited by 1,296
Learning and relearning in Boltzmann machines
Hinton, G. E., & Sejnowski, T. J. (2001) · Graphical Models · doi:10.7551/mitpress/3349.003.0005 · Cited by 1,037
Self-citationReplay and Time Compression of Recurring Spike Sequences in the Hippocampus
Nádasdy, Z., Hirase, H., Czurkó, A., Csicsvari, J., & Buzsáki, G. (1999) · The Journal of Neuroscience · doi:10.1523/jneurosci.19-21-09497.1999 · Cited by 932 · cited 1×
“There is evidence that high-speed, top-down sequence generation occurs during the spindle stage of sleep [Lee and Wilson, 2002, N´adasdy et al., 1999].”
BackgroundLillicrap, T. P., Santoro, A., Marris, L., Akerman, C. J., & Hinton, G. (2020) · Nature Reviews Neuroscience · doi:10.1038/s41583-020-0277-3 · Cited by 908 · cited 3×
“But biologists often object to models that use weight-sharing claiming that there is no obvious way to replicate the weights [Lillicrap et al., 2020].”
BackgroundInfluentialSelf-citationCrick, F., & Mitchison, G. (1983) · Nature · doi:10.1038/304111a0 · Cited by 865
Set Transformer: A Framework for Attention-based Permutation-Invariant Neural Networks
Lee, J., Lee, Y., Kim, J., Kosiorek, A. R., Choi, S., & Teh, Y. W. (2018) · arXiv · doi:10.48550/arxiv.1810.00825 · Cited by 691
Vinyals, O., Kaiser, L., Koo, T., Petrov, S., Sutskever, I., & Hinton, G. (2014) · arXiv · doi:10.48550/arxiv.1412.7449 · Cited by 464
Self-citationExploring Simple Siamese Representation Learning
Chen, X., & He, K. (2020) · arXiv · doi:10.48550/arxiv.2011.10566 · Cited by 381
Some Demonstrations of the Effects of Structural Descriptions in Mental Imagery*
Hinton, G. (1979) · Cognitive Science · doi:10.1207/s15516709cog0303_3 · Cited by 221 · cited 2×
“…psychological evidence that people parse visual scenes into part-whole hierarchies and model the viewpoint-invariant spatial relationship be-tween a part and a whole as the coordinate transformation between intrinsic coordinate frames that they assign to the part and the whole [Hinton, 1979].”
BackgroundSelf-citationKosiorek, A. R., Sabour, S., Teh, Y. W., & Hinton, G. E. (2019) · arXiv · doi:10.48550/arxiv.1906.06818 · Cited by 163
Self-citation
Cited by
Scaling Up Your Kernels to 31×31: Revisiting Large Kernel Design in CNNs
Ding, X., Zhang, X., Han, J., & Ding, G. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01166 · Cited by 1,447
Interpretable machine learning: Fundamental principles and 10 grand challenges
Rudin, C., Chen, C., Chen, Z., Huang, H., Semenova, L., & Zhong, C. (2022) · Statistics Surveys · doi:10.1214/21-ss133 · Cited by 958
Mobile-Former: Bridging MobileNet and Transformer
Chen, Y., Dai, X., Chen, D., Liu, M., Dong, X., Yuan, L., & Liu, Z. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.00520 · Cited by 731
Scaling Vision Transformers to Gigapixel Images via Hierarchical Self-Supervised Learning
Chen, R. J., Chen, C., Li, Y., Chen, T. Y., Trister, A. D., Krishnan, R. G., & Mahmood, F. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01567 · Cited by 621
The neuroconnectionist research programme
Doerig, A., Sommers, R. P., Seeliger, K., Richards, B., Ismael, J., Lindsay, G. W., Kording, K. P., Konkle, T., van Gerven, M. A. J., Kriegeskorte, N., et al. (2023) · Nature Reviews Neuroscience · doi:10.1038/s41583-023-00705-w · Cited by 233
Metadata
- Authors
- Hinton, G.
- Venue
- Neural Computation
- Publisher
- MIT Press
- Published
- 2023-02-17
- Volume
- 35
- Issue
- 3
- Pages
- 413-452
- ISSN
- 0899-7667 (ISSN-L)
- 1530-888X
- OpenAlex
- W3132401450
- Cited by
- 157
- References
- 24
- FWCI
- 9.23
- Type
- journal article
- Language
- en
- Open access
- green
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.