We propose a simple, tractable lower bound on the mutual information contained in the joint generative density of any latent variable generative model: the GILBO (Generative Information Lower BOund). It offers a data independent measure of the complexity of the learned latent variable description, giving the log of the effective description length. It is well-defined for both VAEs and GANs. We ...

We propose a framework for dependency parsing based on a combination of discriminative and generative models. We use a discriminative model to obtain a kbest list of candidate parses, and subsequently rerank those candidates using a generative model. We show how this approach allows us to evaluate a variety of generative models, without needing different parser implementations. Moreover, we pre...

This paper describes how a group working around the emotionally-charged topic of cultural and linguistic appropriateness created new knowledge about practice, and in particular practice about keeping a knowledge problem constantly open for inquiry. The group’s work focused on the creation of a database of materials to support and educate practitioners in their field. Interviews with 23 particip...

There is a close relationship between computational creativity in the arts and generative art. Prior work is summarized here as to how complexity theory can serve as a theoretical context for generative art. This is extended to show how complexity theory can also illuminate discussions regarding computational creativity. Another prior proposal is also summarized. It suggests that a world-view c...

Although Generative Adversarial Networks (GANs) have shown remarkable success in various tasks, they still face challenges in generating high quality images. In this paper, we propose Stacked Generative Adversarial Networks (StackGAN) aimed at generating high-resolution photorealistic images. First, we propose a two-stage generative adversarial network architecture, StackGAN-v1, for textto-imag...

We evolved virtual modular robots for locomotion using a direct and a generative encoding. An important benefit of modular robots in evolutionary robotics is the ability to quickly evaluate various morphologies and control systems in reality. Our use of a modular approach to construct robot morphologies is thus geared towards enabling us to construct robots in reality. However, our approach can...

A large body of research has established that, under relatively simple task conditions, human observers integrate uncertain sensory information with learned prior knowledge in an approximately Bayes-optimal manner. However, in many natural tasks, observers must perform this sensory-plus-prior integration when the underlying generative model of the environment consists of multiple causes. Here w...

Generative modeling, which learns joint probability distribution from training data and generates samples according to it, is an important task in machine learning and artificial intelligence. Inspired by probabilistic interpretation of quantum physics, we propose a generative model using matrix product states, which is a tensor network originally proposed for describing (particularly one-dimen...

What makes a language a natural language? A longstanding tradition in generative grammar holds that a language is natural just in case it is learnable under a constellation of auxiliary assumptions about input evidence available to children. Yet another approach seeks some key mathematical property that distinguishes the natural languages from all possible symbol-systems. With some exceptions f...