Representation learning is concerned with training machine learning algorithms to learn useful representations, e.g. those that are interpretable, have latent features, or can be used for transfer learning.
( Image credit: Visualizing and Understanding Convolutional Networks )
|TREND||DATASET||BEST METHOD||PAPER TITLE||PAPER||CODE||COMPARE|
We therefore propose Cross-View Training (CVT), a semi-supervised learning algorithm that improves the representations of a Bi-LSTM sentence encoder using a mix of labeled and unlabeled data.
Ranked #1 on CCG Supertagging on CCGBank
Specifically, we target semi-supervised classification performance, and we meta-learn an algorithm -- an unsupervised weight update rule -- that produces representations useful for this task.
This paper describes InfoGAN, an information-theoretic extension to the Generative Adversarial Network that is able to learn disentangled representations in a completely unsupervised manner.
Ranked #2 on Image Generation on Stanford Dogs
In recent years, supervised learning with convolutional networks (CNNs) has seen huge adoption in computer vision applications.
Ranked #3 on Image Clustering on Tiny-ImageNet
We also present a detailed empirical analysis of the key factors that are required to achieve these gains, including the trade-offs between (1) positive transfer and capacity dilution and (2) the performance of high and low resource languages at scale.
Many recent methods for unsupervised or self-supervised representation learning train feature extractors by maximizing an estimate of the mutual information (MI) between different views of the data.
The estimation of an f-divergence between two probability distributions based on samples is a fundamental problem in statistics and machine learning.