Sampling from generative models has become a crucial tool for applications like data synthesis and augmentation. Diffusion, Flow Matching and Continuous Normalising Flows have shown effectiveness across various modalities, and rely on latent variables for generation. For experimental design or creative applications that require more control over the generation process, it has become common to manipulate the latent variable directly. However, existing approaches for performing such manipulations (e.g. interpolation or forming low-dimensional representations) only work well in special cases or are network or data-modality specific. We propose Latent Optimal Linear combinations (LOL) as a general-purpose method to form linear combinations of latent variables that adhere to the assumptions of the generative model. As LOL is easy to implement and naturally addresses the broader task of forming any linear combinations, e.g. the construction of subspaces of the latent space, LOL dramatically simplifies the creation of expressive low-dimensional representations of high-dimensional objects.

We analyse Semantic Segmentation Neural Networks running on an autonomous aircraft to estimate its 6DOF pose during landing. We show that automated reasoning techniques from neural network verification can be used to analyse the conditions under which the networks can operate safely, thus providing enhanced assurance guarantees on the behaviour of the overall pose estimation systems.

We demonstrate a unified approach to rigorous design of safety-critical autonomous systems using the VerifAI toolkit for formal analysis of AI-based systems. VerifAI provides an integrated toolchain for tasks spanning the design process, including modeling, falsification, debugging, and ML component retraining. We evaluate all of these applications in an industrial case study on an experimental autonomous aircraft taxiing system developed by Boeing, which uses a neural network to track the centerline of a runway. We define runway scenarios using the Scenic probabilistic programming language, and use them to drive tests in the X-Plane flight simulator. We first perform falsification, automatically finding environment conditions causing the system to violate its specification by deviating significantly from the centerline (or even leaving the runway entirely). Next, we use counterexample analysis to identify distinct failure cases, and confirm their root causes with specialized testing. Finally, we use the results of falsification and debugging to retrain the network, eliminating several failure cases and improving the overall performance of the closed-loop system.

Many decision tools and complex decision systems require components that use learning technology to improve the quality of the decisions, based on observations (such as sensor data). In order to employ these tools and systems in high- or medium-risk applications, the design, implementation, and deployment process needs to follow principled verification, validation, and testing procedures that assure a reliable operation. This task is far from being trivial because of the very nature of learning - a technology that provides tools for making decisions under uncertainty. Only little research efforts have been dedicated so far to validating and testing learning-based systems. This paper describes a novel tool for the testing and the validation of learning systems and a set of statistical tests that are employed by this tool for the assessment of learned classification decisions. We also describe some aspects of the underlying theoretical and experimental framework for the validation and testing of systems that learn.