# Learning to Learn in Simulation

## TL;DR

This paper introduces a deep reinforcement learning approach to train a curiosity-driven agent in simulation for onboard drone object detection, aiming to automate and optimize data collection with minimal human input.

## Contribution

It presents a novel reward function enabling a curiosity agent to balance rapid training and minimal human intervention in a drone environment.

## Key findings

- The curiosity agent successfully learns to train the object detection model efficiently.
- Adjusting the reward function parameter influences the agent's focus on speed versus minimal human input.
- The approach demonstrates potential for automated, real-time training in robotic systems.

## Abstract

Deep learning often requires the manual collection and annotation of a training set. On robotic platforms, can we partially automate this task by training the robot to be curious, i.e., to seek out beneficial training information in the environment? In this work, we address the problem of curiosity as it relates to online, real-time, human-in-the-loop training of an object detection algorithm onboard a drone, where motion is constrained to two dimensions. We use a 3D simulation environment and deep reinforcement learning to train a curiosity agent to, in turn, train the object detection model. This agent could have one of two conflicting objectives: train as quickly as possible, or train with minimal human input. We outline a reward function that allows the curiosity agent to learn either of these objectives, while taking into account some of the physical characteristics of the drone platform on which it is meant to run. In addition, We show that we can weigh the importance of achieving these objectives by adjusting a parameter in the reward function.