# A Useful Motif for Flexible Task Learning in an Embodied Two-Dimensional Visual Environment

## TL;DR

This paper explores how simple modifications to neural network modules can enhance the speed and flexibility of visual task learning in an embodied touchscreen environment, inspired by biological and AI insights.

## Contribution

It demonstrates that minor changes in nonlinear activations within modules significantly improve learning speed and task-switching ability in a simulated visual environment.

## Key findings

- Activation modifications boost learning speed
- Modules adapt better to new tasks
- Simple architectural tweaks have large effects

## Abstract

Animals (especially humans) have an amazing ability to learn new tasks quickly, and switch between them flexibly. How brains support this ability is largely unknown, both neuroscientifically and algorithmically. One reasonable supposition is that modules drawing on an underlying general-purpose sensory representation are dynamically allocated on a per-task basis. Recent results from neuroscience and artificial intelligence suggest the role of the general purpose visual representation may be played by a deep convolutional neural network, and give some clues how task modules based on such a representation might be discovered and constructed. In this work, we investigate module architectures in an embodied two-dimensional touchscreen environment, in which an agent's learning must occur via interactions with an environment that emits images and rewards, and accepts touches as input. This environment is designed to capture the physical structure of the task environments that are commonly deployed in visual neuroscience and psychophysics. We show that in this context, very simple changes in the nonlinear activations used by such a module can significantly influence how fast it is at learning visual tasks and how suitable it is for switching to new tasks.