# Deep Reinforcement Learning on a Budget: 3D Control and Reasoning Without a Supercomputer

## TL;DR

This paper introduces a suite of complex reasoning tasks in 3D environments that can be trained on mid-range hardware within 24 hours, making deep reinforcement learning research more accessible and less reliant on supercomputers.

## Contribution

It presents a new set of challenging 3D reasoning tasks using a fast, efficient environment (ViZDoom) and provides baseline agents that can be trained on consumer hardware, lowering barriers to research.

## Key findings

- High-speed simulation (12000fps) enables quick training.
- Scenarios with adjustable difficulty highlight current algorithm limitations.
- Baseline agents can be trained within 24 hours on mid-range hardware.

## Abstract

An important goal of research in Deep Reinforcement Learning in mobile robotics is to train agents capable of solving complex tasks, which require a high level of scene understanding and reasoning from an egocentric perspective. When trained from simulations, optimal environments should satisfy a currently unobtainable combination of high-fidelity photographic observations, massive amounts of different environment configurations and fast simulation speeds. In this paper we argue that research on training agents capable of complex reasoning can be simplified by decoupling from the requirement of high fidelity photographic observations. We present a suite of tasks requiring complex reasoning and exploration in continuous, partially observable 3D environments. The objective is to provide challenging scenarios and a robust baseline agent architecture that can be trained on mid-range consumer hardware in under 24h. Our scenarios combine two key advantages: (i) they are based on a simple but highly efficient 3D environment (ViZDoom) which allows high speed simulation (12000fps); (ii) the scenarios provide the user with a range of difficulty settings, in order to identify the limitations of current state of the art algorithms and network architectures. We aim to increase accessibility to the field of Deep-RL by providing baselines for challenging scenarios where new ideas can be iterated on quickly. We argue that the community should be able to address challenging problems in reasoning of mobile agents without the need for a large compute infrastructure.