Binding via Reconstruction Clustering

By Klaus Greff, Rupesh Kumar Srivastava, J. Urgen Schmidhuber. 2015

42 citations recorded by the source corpus.

Disentangled distributed representations of data are desirable for machine learning, since they are more expressive and can generalize from fewer examples. However, for complex data, the distributed representations of multiple objects present in the same input can interfere and lead to ambiguities, which is commonly referred to as the binding problem. We argue for the importance of the binding problem to the field of representation learning, and develop a probabilistic framework that explicitly models inputs as a composition of multiple objects. We propose an unsupervised algorithm that uses denoising autoencoders to dynamically bind features together in multi-object inputs through an Expectation-Maximization-like clustering process. The effectiveness of this method is demonstrated on artificially generated datasets of binary images, showing that it can even generalize to bind together new objects never seen by the autoencoder during training.

Subfields: Biophysics, 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 Biophysics

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

Invariant Information Clustering for Unsupervised Image Classification and Segmentation (2018, methodology): “DAC [8], JULE [52], DeepCluster [7], ADC [24] and DEC [51] rely on the inherent visual consistency and disentangling properties [23] of CNNs to produce cluster assignments, which are processed and reinforced in each iteration.”
Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation (2016, background): “Unsupervised detection of objects in visual scenes is an open problem in computer vision, although there has been recent progress in obtaining objects directly from image or motion data [9, 8, 11].”
Tagger: Deep Unsupervised Perceptual Grouping (2016, background): “(a) Convergence of Tagger over iterative inference AMI RC [8] 0.”
Neural Expectation Maximization (2017, methodology): “Consistent with earlier work [8, 7], we evaluate the quality of the learned groupings with respect to the ground truth while ignoring the background and overlap regions.”
Deep Successor Reinforcement Learning (2016, background): “In order to scale up our approach to more expressive environments, it will be crucial to combine various deep generative and self-supervised models [6, 9, 10, 12, 16, 25, 27, 40] with our approach.”
Invariant Information Distillation for Unsupervised Image Segmentation and Clustering (2018, background): “Greff et al. [2016] is a generative approach; a function is trained to denoise its input using several clusters to distribute the representation.”
Generative Modeling of Infinite Occluded Objects for Compositional Scene Representation (2019, methodology): “Reconstruction Clustering (RC) (Greff et al., 2016b) iteratively masks the input image with posterior probabilities and reconstructs the image in a compositional manner with a pretrained denoising autoencoder.”
Entity Abstraction in Visual Model-Based Reinforcement Learning (2019, background): “OP3 builds upon recently proposed ideas in grounding entity representations via inference on a symmetrically factorized generative model of static [8, 17, 42] and dynamic [43] scenes, whose advantage over other methods for grounding [5, 44–47] is the ability to refine the grounding with new informat”
Recurrent Ladder Networks (2017, methodology): “We assessed the segmentation quality by the AMI score which was computed similarly to [13, 12] ignoring the background in the case of a uniform zero-valued background and overlap regions where different objects have the same color.”