# Beyond imitation: Zero-shot task transfer on robots by learning concepts as cognitive programs

## TL;DR

This paper presents a computational framework enabling robots to learn, infer, and transfer high-level concepts as programs, significantly enhancing their ability to understand and perform tasks in varied environments by mimicking human-like concept learning.

## Contribution

It introduces a novel cognitive computer architecture with a specialized instruction set for concept learning, including imagination and recursion, facilitating zero-shot task transfer in robots.

## Key findings

- Robots can interpret novel schematic concepts.
- Hierarchical abstractions simplify learning complex concepts.
- Robots successfully apply learned concepts in different scenarios.

## Abstract

Humans can infer concepts from image pairs and apply those in the physical world in a completely different setting, enabling tasks like IKEA assembly from diagrams. If robots could represent and infer high-level concepts, it would significantly improve their ability to understand our intent and to transfer tasks between different environments. To that end, we introduce a computational framework that replicates aspects of human concept learning. Concepts are represented as programs on a novel computer architecture consisting of a visual perception system, working memory, and action controller. The instruction set of this "cognitive computer" has commands for parsing a visual scene, directing gaze and attention, imagining new objects, manipulating the contents of a visual working memory, and controlling arm movement. Inferring a concept corresponds to inducing a program that can transform the input to the output. Some concepts require the use of imagination and recursion. Previously learned concepts simplify the learning of subsequent more elaborate concepts, and create a hierarchy of abstractions. We demonstrate how a robot can use these abstractions to interpret novel concepts presented to it as schematic images, and then apply those concepts in dramatically different situations. By bringing cognitive science ideas on mental imagery, perceptual symbols, embodied cognition, and deictic mechanisms into the realm of machine learning, our work brings us closer to the goal of building robots that have interpretable representations and commonsense.