A Procedural World Generation Framework for Systematic Evaluation of Continual Learning

TL;DR

This paper introduces a procedural world generation framework that creates customizable, continuous urban scene data streams to enable systematic evaluation of continual learning methods in dynamic environments.

Contribution

It presents a novel, modular simulation framework for generating diverse, controllable datasets to facilitate comprehensive analysis of continual learning algorithms.

Findings

Enables detailed analysis of continual learning schemes.

Facilitates systematic evaluation with customizable data streams.

Supports real-time, endless urban scene generation.

Abstract

Several families of continual learning techniques have been proposed to alleviate catastrophic interference in deep neural network training on non-stationary data. However, a comprehensive comparison and analysis of limitations remains largely open due to the inaccessibility to suitable datasets. Empirical examination not only varies immensely between individual works, it further currently relies on contrived composition of benchmarks through subdivision and concatenation of various prevalent static vision datasets. In this work, our goal is to bridge this gap by introducing a computer graphics simulation framework that repeatedly renders only upcoming urban scene fragments in an endless real-time procedural world generation process. At its core lies a modular parametric generative model with adaptable generative factors. The latter can be used to flexibly compose data streams, which significantly facilitates a detailed analysis and allows for effortless investigation of various continual learning schemes.