The Global Latent Workspace: towards AI models of flexible cognition

Introduction

Recent advances in deep learning have allowed Artificial Intelligence (AI) to reach human-level performance in many sensory, perceptual, linguistic, or cognitive tasks. There is a growing need, however, for novel, brain-inspired cognitive architectures to achieve more robust and flexible cognition.

Global Workspace Theory

The Global Workspace Theory refers to a large-scale system integrating and distributing information among networks of specialized modules to create higher-level forms of cognition and awareness. It is one of the dominant neuroscientific accounts of higher-level brain function.

Proposed Implementation

We argue that the time is ripe to consider explicit implementations of this theory using deep learning techniques. We propose a roadmap based on unsupervised neural translation between multiple latent spaces:

1. Neural networks trained for distinct tasks
2. Networks trained on distinct sensory inputs and/or modalities

This approach aims to create a unique, amodal global latent workspace (GLW).

Semantic Grounding

Sensory inputs that are broadcast in this GLW acquire meaning by connecting them to (or translating them into) the relevant semantic knowledge and representations from language, memory, or decision systems: the semantic grounding property. Conversely, language and semantic representations are grounded in the sensory environment via the same broadcast/translation operation.

Motor Affordances and Embodied Cognition

Finally, broadcasting sensory and semantic inputs to the relevant effector domains can create motor affordances and support embodied cognition. Together, the grounding and affordance properties infuse meaning in AI processes, which can then be combined sequentially (via attentional selection) to enable flexible cognitive functions, i.e., System-2 AI.

Project Goals

The interdisciplinary project will directly implement the GLW framework in deep learning models of growing complexity and evaluate their correspondence with brain networks. It will provide an explicit evaluation of the Global Workspace Theory and push the limits of current deep learning systems towards next-generation AI.