Split-Brain Autoencoders: Unsupervised Learning by Cross-Channel Prediction

Abstract

We propose split-brain autoencoders, a straightforward modification of the traditional autoencoder architecture, for unsupervised representation learning. The method adds a split to the network, resulting in two disjoint sub-networks. Each sub-network is trained to perform a difficult task -- predicting one subset of the data channels from another. Together, the sub-networks extract features from the entire input signal. By forcing the network to solve cross-channel prediction tasks, we induce a representation within the network which transfers well to other, unseen tasks. This method achieves state-of-the-art performance on several large-scale transfer learning benchmarks.

Accepted to CVPR 2017

References in corpus (6)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Generative Adversarial Text to Image Synthesis
• Conditional Image Generation with PixelCNN Decoders
• A Survey on Multi-view Learning
• Density estimation using Real NVP
• Adversarially Learned Inference

Cited by in corpus (11)

• Unsupervised Representation Learning by Predicting Image Rotations
• Contrastive Multiview Coding
• Deep Clustering for Unsupervised Learning of Visual Features
• PixelNet: Representation of the pixels, by the pixels, and for the pixels
• Colorization as a Proxy Task for Visual Understanding
• Time-Contrastive Networks: Self-Supervised Learning from Video
• Unsupervised Representation Learning by Sorting Sequences
• Self-Supervised Feature Learning by Learning to Spot Artifacts
• Mix-and-Match Tuning for Self-Supervised Semantic Segmentation
• Cross-Domain Self-supervised Multi-task Feature Learning using Synthetic Imagery
• Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks