Abstract:In recent years, the remarkable progress of Machine Learning (ML) technologies within the domain of Artificial Intelligence (AI) systems has presented unprecedented opportunities for the aviation industry, paving the way for further advancements in automation, including the potential for single pilot or fully autonomous operation of large commercial airplanes. However, ML technology faces major incompatibilities with existing airborne certification standards, such as ML model traceability and explainability issues or the inadequacy of traditional coverage metrics. Certification of ML-based airborne systems using current standards is problematic due to these challenges. This paper presents a case study of an airborne system utilizing a Deep Neural Network (DNN) for airport sign detection and classification. Building upon our previous work, which demonstrates compliance with Design Assurance Level (DAL) D, we upgrade the system to meet the more stringent requirements of Design Assurance Level C. To achieve DAL C, we employ an established architectural mitigation technique involving two redundant and dissimilar Deep Neural Networks. The application of novel ML-specific data management techniques further enhances this approach. This work is intended to illustrate how the certification challenges of ML-based systems can be addressed for medium criticality airborne applications.
Abstract:Neuromodulatory receptors in presynaptic position have the ability to suppress synaptic transmission for seconds to minutes when fully engaged. This effectively alters the synaptic strength of a connection. Much work on neuromodulation has rested on the assumption that these effects are uniform at every neuron. However, there is considerable evidence to suggest that presynaptic regulation may be in effect synapse-specific. This would define a second "weight modulation" matrix, which reflects presynaptic receptor efficacy at a given site. Here we explore functional consequences of this hypothesis. By analyzing and comparing the weight matrices of networks trained on different aspects of a task, we identify the potential for a low complexity "modulation matrix", which allows to switch between differently trained subtasks while retaining general performance characteristics for the task. This means that a given network can adapt itself to different task demands by regulating its release of neuromodulators. Specifically, we suggest that (a) a network can provide optimized responses for related classification tasks without the need to train entirely separate networks and (b) a network can blend a "memory mode" which aims at reproducing memorized patterns and a "novelty mode" which aims to facilitate classification of new patterns. We relate this work to the known effects of neuromodulators on brain-state dependent processing.