Consciousness Prior

Experiment on a proposal for a new kind of prior for top-level abstract representations, inspired by our understanding of consciousness.

Start date: November 2017
Category: Fundamental Research
Contact point: William Fedus (liam.fedus@gmail.com)

Abstract

Experiments are proposed to explore a novel prior for representation learning, which can be combined with other priors in order to help disentangling abstract factors from each other. It is inspired by the phenomenon of consciousness seen as the formation of a low-dimensional combination of a few concepts constituting a conscious thought, i.e., consciousness as awareness at a particular time instant. This provides a powerful constraint on the representation in that such low-dimensional thought vectors can correspond to statements about reality which are true, highly probable, or very useful for taking decisions. Instead of making predictions in the sensory (e.g. pixel) space, the consciousness prior allows the agent to make predictions in the abstract space, with only a few dimensions of that space being involved in each of these predictions. Experiments on a synthetic dataset are proposed to validate some of the mechanisms proposed to implement the consciousness prior, in the simplest scenario where the consciousness mechanism is only used to make a prediction.

Access the full proposal

Resources