AutoJoin: Efficient Adversarial Training against Gradient-Free Perturbations for Robust Maneuvering via Denoising Autoencoder and Joint Learning
This addresses the need for trustworthy and efficient adversarial training in machine learning systems for maneuvering, representing a strong specific gain rather than a foundational advancement.
The paper tackled the problem of improving robustness in perception-to-control systems for image-based maneuvering by proposing AutoJoin, a gradient-free adversarial training technique that achieved up to 40% performance increases against perturbations and up to 300% improvements in clean performance while saving up to 86% time per training epoch and 90% training data.
With the growing use of machine learning algorithms and ubiquitous sensors, many `perception-to-control' systems are being developed and deployed. To ensure their trustworthiness, improving their robustness through adversarial training is one potential approach. We propose a gradient-free adversarial training technique, named AutoJoin, to effectively and efficiently produce robust models for image-based maneuvering. Compared to other state-of-the-art methods with testing on over 5M images, AutoJoin achieves significant performance increases up to the 40% range against perturbations while improving on clean performance up to 300%. AutoJoin is also highly efficient, saving up to 86% time per training epoch and 90% training data over other state-of-the-art techniques. The core idea of AutoJoin is to use a decoder attachment to the original regression model creating a denoising autoencoder within the architecture. This architecture allows the tasks `maneuvering' and `denoising sensor input' to be jointly learnt and reinforce each other's performance.