Transcription of Neural Discrete Representation Learning
{{id}} {{{paragraph}}}
Example: biology
Neural Discrete Representation Learning
Neural Discrete Representation LearningAaron van den useful representations without supervision remains a key challenge inmachine Learning . In this paper, we propose a simple yet powerful generativemodel that learns such Discrete representations. Our model, the Vector Quantised-Variational AutoEncoder (VQ-VAE), differs from VAEs in two key ways: theencoder network outputs Discrete , rather than continuous, codes; and the prioris learnt rather than static. In order to learn a Discrete latent Representation , weincorporate ideas from vector quantisation (VQ).
The representation z e(x) is passed through the discretisation bottleneck followed by mapping onto the nearest element of embedding eas given in equations 1 and 2. z q(x) = e k; where k= argmin j kz e(x) e jk 2 (2) 3.2 Learning Note that there is no real gradient defined for equation 2, however we approximate the gradient
Loading..
Tags:
Information
Domain:
Source:
Link to this page:
Please notify us if you found a problem with this document:
Related search queries
1 Information Representation and Retrieval: An, Information representation and retrieval, Information, 1 REPRESENTATION, Representation, Representation Learning, Graph, NOTICE OF LIMITED SCOPE REPRESENTATION, California, Communication with Clients Rule Approved, Conflict of Interest: Current Clients