One-Class Domain Adaptation via Meta-Learning

TL;DR

This paper introduces a meta-learning approach, OC-DA MAML, for one-class domain adaptation in sensor-based anomaly detection, enabling rapid transfer to new environments with limited labeled data.

Contribution

It extends meta-learning to the one-class domain adaptation setting and proposes a novel task sampling strategy for improved adaptation performance.

Findings

OC-DA MAML outperforms standard MAML on Rainbow-MNIST benchmark.

The method significantly improves target domain performance in real-world sensor data.

Theoretical analysis confirms rapid adaptation capabilities of the proposed approach.

Abstract

The deployment of IoT (Internet of Things) sensor-based machine learning models in industrial systems for anomaly classification tasks poses significant challenges due to distribution shifts, as the training data acquired in controlled laboratory settings may significantly differ from real-time data in production environments. Furthermore, many real-world applications cannot provide a substantial number of labeled examples for each anomalous class in every new environment. It is therefore crucial to develop adaptable machine learning models that can be effectively transferred from one environment to another, enabling rapid adaptation using normal operational data. We extended this problem setting to an arbitrary classification task and formulated the one-class domain adaptation (OC-DA) problem setting. We took a meta-learning approach to tackle the challenge of OC-DA, and proposed a task sampling strategy to adapt any bi-level meta-learning algorithm to OC-DA. We modified the well-established model-agnostic meta-learning (MAML) algorithm and introduced the OC-DA MAML algorithm. We provided a theoretical analysis showing that OC-DA MAML optimizes for meta-parameters that enable rapid one-class adaptation across domains. The OC-DA MAML algorithm is evaluated on the Rainbow-MNIST meta-learning benchmark and on a real-world dataset of vibration-based sensor readings. The results show that OC-DA MAML significantly improves the performance on the target domains and outperforms MAML using the standard task sampling strategy.