Disentangling Factors of Variation via Generative Entangling

Abstract

Here we propose a novel model family with the objective of learning to disentangle the factors of variation in data. Our approach is based on the spike-and-slab restricted Boltzmann machine which we generalize to include higher-order interactions among multiple latent variables. Seen from a generative perspective, the multiplicative interactions emulates the entangling of factors of variation. Inference in the model can be seen as disentangling these generative factors. Unlike previous attempts at disentangling latent factors, the proposed model is trained using no supervised information regarding the latent factors. We apply our model to the task of facial expression classification.

Cited by in corpus (29)

• Deep Convolutional Inverse Graphics Network
• Disentangling by Factorising
• Isolating Sources of Disentanglement in Variational Autoencoders
• DARLA: Improving Zero-Shot Transfer in Reinforcement Learning
• Early Visual Concept Learning with Unsupervised Deep Learning
• Deep Learning of Representations: Looking Forward
• Variational Inference of Disentangled Latent Concepts from Unlabeled Observations
• A Survey of Inductive Biases for Factorial Representation-Learning
• Generalization in anti-causal learning
• Disentangled and Controllable Face Image Generation via 3D Imitative-Contrastive Learning
• Learning Disentangled Representations with Reference-Based Variational Autoencoders
• Towards Visually Explaining Variational Autoencoders
• Towards Privacy-Preserving Visual Recognition via Adversarial Training: A Pilot Study
• "Mental Rotation" by Optimizing Transforming Distance
• Understanding Instance-based Interpretability of Variational Auto-Encoders
• Disentangled Variational Representation for Heterogeneous Face Recognition
• Disentangling Factors of Variation with Cycle-Consistent Variational Auto-Encoders
• Unsupervised Landmark Learning from Unpaired Data
• Disentangled Representations in Neural Models
• Delving into Robust Object Detection from Unmanned Aerial Vehicles: A Deep Nuisance Disentanglement Approach
• Measuring the Biases and Effectiveness of Content-Style Disentanglement
• Independence Promoted Graph Disentangled Networks
• Hyperprior Induced Unsupervised Disentanglement of Latent Representations
• Adversarial Disentanglement with Grouped Observations
• WeLa-VAE: Learning Alternative Disentangled Representations Using Weak Labels
• Unsupervised Disentanglement of Linear-Encoded Facial Semantics
• Disentanglement Learning via Topology
• Set-Based Face Recognition Beyond Disentanglement: Burstiness Suppression With Variance Vocabulary
• Towards Better Understanding of Disentangled Representations via Mutual Information