Representation Learning: A Review and New Perspectives
By Yoshua Bengio, Aaron C. Courville, Pascal Vincent. 2012
14,067 citations recorded by the source corpus.
The success of machine learning algorithms generally depends on data representation, and we hypothesize that this is because different representations can entangle and hide more or less the different explanatory factors of variation behind the data. Although specific domain knowledge can be used to help design representations, learning with generic priors can also be used, and the quest for AI is motivating the design of more powerful representation-learning algorithms implementing such priors. This paper reviews recent work in the area of unsupervised feature learning and deep learning, covering advances in probabilistic models, autoencoders, manifold learning, and deep networks. This motivates longer term unanswered questions about the appropriate objectives for learning good representations, for computing representations (i.e., inference), and the geometrical connections between representation learning, density estimation, and manifold learning.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Yoshua Bengio
- Mohit Bansal
- Sergey Levine
- Jianfeng Gao
- Quoc V. Le
- Percy Liang
- Ruslan Salakhutdinov
- Dacheng Tao
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Dacheng Tao
- Trevor Darrell
- Ming-Hsuan Yang
- Xiaogang Wang
- Jitendra Malik
- Andrew Zisserman
- Yi Yang
In Representation & Self-Supervised Learning
- Yoshua Bengio
- Trevor Darrell
- Dacheng Tao
- Bernhard Schölkopf
- Andrea Vedaldi
- Sergey Levine
- Ishan Misra
- Ruslan Salakhutdinov
Citations in the corpus
- Highway Networks (2015, background): “As argued by Bengio et al. (2013), the use of deep networks can offer both computational and statistical efficiency for complex tasks.”
- FitNets: Hints for Thin Deep Nets (2014, background): “Depth is a fundamental aspect of representation learning, since it encourages the re-use of features, and leads to more abstract and invariant representations at higher layers (Bengio et al., 2013).”
- Network In Network (2013, background): “By abstraction we mean that the feature is invariant to the variants of the same concept [2].”
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation (2013, neutral): “[3] for an excellent survey.”
- Auto-Encoding Variational Bayes (2013, positive): “However, it is well known that this reconstruction criterion is in itself not sufficient for learning useful representations [BCV13].”
- InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets (2016, background): “A popular framework for unsupervised learning is that of representation learning [1, 2], whose goal is to use unlabelled data to learn a representation that exposes important semantic features as easily decodable factors.”
- A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark (2019, background): “Improving sample efficiency has been approached from many angles: few-shot learning (Fei-Fei et al., 2006), transfer learning (Pan & Yang, 2009), domain adaptation (Wang & Deng, 2018), and representation learning (Bengio et al., 2013).”
- A Sober Look at the Unsupervised Learning of Disentangled Representations and their Evaluation (2020, background): “They should contain all the information present in x in a compact and interpretable structure (Bengio et al., 2013; Kulkarni et al., 2015; Chen et al., 2016) while being independent from the task at hand (Goodfellow et al., 2009; Lenc and Vedaldi, 2015).”
- Contrastive Multiview Coding (2019, background): “Unsupervised representation learning is about learning transformations of the data that make subsequent problem solving easier [7].”
- CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning (2016, methodology): “We plan to use CLEVR to study models with explicit short-term memory, facilitating comparisons between values [13, 18, 39, 42]; explore approaches that encourage learning disentangled representations [5]; and investigate methods that compile custom network architectures for different patterns of rea”
- Challenges in representation learning: A report on three machine learning contests (2013, background): “” The purpose of the workshop, organized by Ian Goodfellow, Dumitru Erhan, and Yoshua Bengio, was to explore the latest developments in representation learning, with a special emphasis on testing the capabilities of current representation learning algorithms (See [1] for a recent review) and pushing”
- Learning Features by Watching Objects Move (2016, neutral): “[4] provide an excellent survey.”
- Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzles (2016, background): “There is a rich literature in unsupervised learning of visual representations [5].”
- Colorful Image Colorization (2016, background): “The idea of learning feature representations in this way goes back at least to autoencoders [6].”
- BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling (2019, background): “This distribution most commonly lies on a low-dimensional manifold that can provide insights into the internal representation of the data [1].”
- GSNs : Generative Stochastic Networks (2015, methodology): “The following theorem is proved by Bengio et al. (2013c).”
- Diagnosing and Enhancing VAE Models (2019, background): “In fact, the very utility of generative models of continuous data, and their attendant low-dimensional representations, often hinges on this assumption (Bengio et al., 2013).”
- High Fidelity Visualization of What Your Self-Supervised Representation Knows About (2021, background): “, 2014), that are traditionally put under the broad umbrella term of unsupervised learning (Bengio et al., 2013).”