Continuous Adaptation via Meta-Learning in Nonstationary and Competitive Environments

Introduction

Typically, the basic goal of machine learning is to train a model to perform a task. In Meta-learning, the goal is to train a model to perform the task of training a model to perform a task. Hence in this case the term "Meta-Learning" has the exact meaning you would expect; the word "Meta" has the precise function of introducing a layer of abstraction.

The meta-learning task can be made more concrete by a simple example. Consider the CIFAR-100 classification task that we used for our data competition. We can alter this task from being a 100-class classification problem to a collection of 100 binary classification problems. The goal of Meta-Learning here is to design and train and single binary classifier that will perform well on a randomly sampled task given a limited amount of training data for that specific task. In other words, we would like to train a model of perform the following procedure.

- A task is sampled. The task is "Is X a dog?" - A small set of labeled training data is provided to the model. The labels represent whether or not the image is a dog. - The model uses the training data to adjust itself to the specific task of checking whether or not an image is a picture of a dog.

In this paper, a probabilistic framework for meta learning is derived then applied to tasks involving simulated robotic spiders. This framework generalizes the typical machine learning set up using Markov Decision Processes.