Trojan Horse Hunt in Time Series Forecasting for Space Operations

TL;DR

This paper introduces a competition focused on detecting and reconstructing adversarial triggers in satellite telemetry forecasting models, highlighting the importance of AI security in space operations and other safety-critical domains.

Contribution

It presents a novel challenge for identifying triggers in poisoned time series models, providing datasets, baseline methods, and fostering research on AI security in space applications.

Findings

Baseline neural cleanse method is ineffective for time series triggers.

Participants need to develop new approaches for trigger detection.

The competition emphasizes AI security in safety-critical systems.

Abstract

This competition hosted on Kaggle (https://www.kaggle.com/competitions/trojan-horse-hunt-in-space) is the first part of a series of follow-up competitions and hackathons related to the "Assurance for Space Domain AI Applications" project funded by the European Space Agency (https://assurance-ai.space-codev.org/). The competition idea is based on one of the real-life AI security threats identified within the project -- the adversarial poisoning of continuously fine-tuned satellite telemetry forecasting models. The task is to develop methods for finding and reconstructing triggers (trojans) in advanced models for satellite telemetry forecasting used in safety-critical space operations. Participants are provided with 1) a large public dataset of real-life multivariate satellite telemetry (without triggers), 2) a reference model trained on the clean data, 3) a set of poisoned neural hierarchical interpolation (N-HiTS) models for time series forecasting trained on the dataset with injected triggers, and 4) Jupyter notebook with the training pipeline and baseline algorithm (the latter will be published in the last month of the competition). The main task of the competition is to reconstruct a set of 45 triggers (i.e., short multivariate time series segments) injected into the training data of the corresponding set of 45 poisoned models. The exact characteristics (i.e., shape, amplitude, and duration) of these triggers must be identified by participants. The popular Neural Cleanse method is adopted as a baseline, but it is not designed for time series analysis and new approaches are necessary for the task. The impact of the competition is not limited to the space domain, but also to many other safety-critical applications of advanced time series analysis where model poisoning may lead to serious consequences.