arXiv subjects

Abanoub Ghobrial

Publications and source records attributed to Abanoub Ghobrial.

7 recordsLinked to original sources

arXiv2026

DIRA-SS:Dynamic Domain Incremental Regularised Adaptation -- Self-Supervised

Autonomous systems (AS) often rely on Deep Neural Network (DNN) classifiers to operate in complex and dynamically changing environments. However, during operation, these classifiers may encounter domains that differ from those seen during development, causing performance degradation under distribution shift. Removing systems from operation for labelled data collection and retraining is often impractical, particularly when adaptation must occur quickly and at scale. This paper introduces DIRA-SS, a self-supervised extension of Dynamic Incremental Regularised Adaptation (DIRA) that enables online domain adaptation using only a small number of unlabelled target-domain samples. DIRA-SS augments an existing classifier with an auxiliary retraining branch and adapts the shared feature representation through a rotation-prediction task, while elastic weight consolidation regularises important source-domain parameters to reduce destructive updates. This allows the model to benefit from transfer learning without requiring classification labels during operation. We evaluate DIRA-SS on CIFAR-10C, CIFAR-100C, and ImageNet-C using ResNet architectures under severe common corruptions. The results show that DIRA-SS substantially improves performance over the non-adapted source model, achieves accuracy close to the supervised DIRA method, and outperforms existing unsupervised test-time adaptation baselines on ImageNet-C when using only 100 target-domain samples.

cs.LG↗

arXiv2024

DIRA: Dynamic Domain Incremental Regularised Adaptation

Autonomous systems (AS) often use Deep Neural Network (DNN) classifiers to allow them to operate in complex, high-dimensional, non-linear, and dynamically changing environments. Due to the complexity of these environments, DNN classifiers may output misclassifications during operation when they face domains not identified during development. Removing a system from operation for retraining becomes impractical as the number of such AS increases. To increase AS reliability and overcome this limitation, DNN classifiers need to have the ability to adapt during operation when faced with different operational domains using a few samples (e.g. 2 to 100 samples). However, retraining DNNs on a few samples is known to cause catastrophic forgetting and poor generalisation. In this paper, we introduce Dynamic Incremental Regularised Adaptation (DIRA), an approach for dynamic operational domain adaption of DNNs using regularisation techniques. We show that DIRA improves on the problem of forgetting and achieves strong gains in performance when retraining using a few samples from the target domain. Our approach shows improvements on different image classification benchmarks aimed at evaluating robustness to distribution shifts (e.g.CIFAR-10C/100C, ImageNet-C), and produces state-of-the-art performance in comparison with other methods from the literature.

cs.LG↗

arXiv2023

Evaluation Metrics for DNNs Compression

There is a lot of ongoing research effort into developing different techniques for neural networks compression. However, the community lacks standardised evaluation metrics, which are key to identifying the most suitable compression technique for different applications. This paper reviews existing neural network compression evaluation metrics and implements them into a standardisation framework called NetZIP. We introduce two novel metrics to cover existing gaps of evaluation in the literature: 1) Compression and Hardware Agnostic Theoretical Speed (CHATS) and 2) Overall Compression Success (OCS). We demonstrate the use of NetZIP using two case studies on two different hardware platforms (a PC and a Raspberry Pi 4) focusing on object classification and object detection.

cs.LG↗

arXiv2023

A Trustworthiness Score to Evaluate DNN Predictions

Due to the black box nature of deep neural networks (DNN), the continuous validation of DNN during operation is challenging with the absence of a human monitor. As a result this makes it difficult for developers and regulators to gain confidence in the deployment of autonomous systems employing DNN. It is critical for safety during operation to know when DNN's predictions are trustworthy or suspicious. With the absence of a human monitor, the basic approach is to use the model's output confidence score to assess if predictions are trustworthy or suspicious. However, the model's confidence score is a result of computations coming from a black box, therefore lacks transparency and makes it challenging to automatedly credit trustworthiness to predictions. We introduce the trustworthiness score (TS), a simple metric that provides a more transparent and effective way of providing confidence in DNN predictions compared to model's confidence score. The metric quantifies the trustworthiness in a prediction by checking for the existence of certain features in the predictions made by the DNN. We also use the underlying idea of the TS metric, to provide a suspiciousness score (SS) in the overall input frame to help in the detection of suspicious frames where false negatives exist. We conduct a case study using YOLOv5 on persons detection to demonstrate our method and usage of TS and SS. The case study shows that using our method consistently improves the precision of predictions compared to relying on model confidence score alone, for both 1) approving of trustworthy predictions (~20% improvement) and 2) detecting suspicious frames (~5% improvement).

cs.LG↗

arXiv2022

Safety Validation of Autonomous Vehicles using Assertion Checking

Safety and mission performance validation of autonomous vehicles (AVs) is a major challenge. In this paper we describe a methodology for constructing and applying assertion checks to validate the behaviour of an AV operating either in simulation or in the real world. We have identified a taxonomy of assertion types and the general format of their specification, and we have developed procedures for translating driving codes of practice to yield formal logical expressions that can be monitored automatically by computer, either by direct translation or by physical modelling. We have developed examples of assertions derived from the UK Highway Code (UKHC), as an example of a code of practice. We illustrate the approach with an example of assertion checking for vehicle overtaking, using a geospatial information system in an SQL database for validation and performance assessment. We present initial simulation and runtime monitoring experiments that apply assertions relevant in this overtaking scenario together with an analysis of the safety and mission performance characteristics measured.

cs.DB↗

arXiv2022

On Determinism of Game Engines used for Simulation-based Autonomous Vehicle Verification

Game engines are increasingly used as simulation platforms by the autonomous vehicle (AV) community to develop vehicle control systems and test environments. A key requirement for simulation-based development and verification is determinism, since a deterministic process will always produce the same output given the same initial conditions and event history. Thus, in a deterministic simulation environment, tests are rendered repeatable and yield simulation results that are trustworthy and straightforward to debug. However, game engines are seldom deterministic. This paper reviews and identifies the potential causes of non-deterministic behaviours in game engines. A case study using CARLA, an open-source autonomous driving simulation environment powered by Unreal Engine, is presented to highlight its inherent shortcomings in providing sufficient precision in experimental results. Different configurations and utilisations of the software and hardware are explored to determine an operational domain where the simulation precision is sufficiently low i.e.\ variance between repeated executions becomes negligible for development and testing work. Finally, a method of a general nature is proposed, that can be used to find the domains of permissible variance in game engine simulations for any given system configuration.

cs.RO↗

arXiv2020

An Agency-Directed Approach to Test Generation for Simulation-based Autonomous Vehicle Verification

Simulation-based verification is beneficial for assessing otherwise dangerous or costly on-road testing of autonomous vehicles (AV). This paper addresses the challenge of efficiently generating effective tests for simulation-based AV verification using software testing agents. The multi-agent system (MAS) programming paradigm offers rational agency, causality and strategic planning between multiple agents. We exploit these aspects for test generation, focusing in particular on the generation of tests that trigger the precondition of an assertion. On the example of a key assertion we show that, by encoding a variety of different behaviours respondent to the agent's perceptions of the test environment, the agency-directed approach generates twice as many effective tests than pseudo-random test generation, while being both efficient and robust. Moreover, agents can be encoded to behave naturally without compromising the effectiveness of test generation. Our results suggest that generating tests using agency-directed testing significantly improves upon random and simultaneously provides more realistic driving scenarios.

cs.MA↗