Imagination Technologies

Contract-based design is a promising methodology for taming the complexity of developing sophisticated systems. A formal contract distinguishes between assumptions, which are constraints that the designer of a component puts on the environments in which the component can be used safely, and guarantees, which are promises that the designer asks from the team that implements the component. A theory of formal contracts can be formalized as an interface theory, which supports the composition and refinement of both assumptions and guarantees. Although there is a rich landscape of contract-based design methods that address functional and extra-functional properties, we present the first interface theory designed to ensure system-wide security properties. Our framework provides a refinement relation and a composition operation that support both incremental design and independent implementability. We develop our theory for both stateless and stateful interfaces. Additionally, we introduce information-flow contracts where assumptions and guarantees are sets of flow relations. We use these contracts to illustrate how to enrich information-flow interfaces with a semantic view. We illustrate the applicability of our framework with two examples inspired by the automotive domain.

Building a real-time spatio-temporal forecasting system is a challenging problem with many practical applications such as traffic and road network management. Most forecasting research focuses on achieving (often marginal) improvements in evaluation metrics such as MAE/MAPE on static benchmark datasets, with less attention paid to building practical pipelines which achieve timely and accurate forecasts when the network is under heavy load. Transport authorities also need to leverage dynamic data sources such as roadworks and vehicle-level flow data, while also supporting ad-hoc inference workloads at low cost. Our cloud-based forecasting solution Foresight, developed in collaboration with Transport for the West Midlands (TfWM), is able to ingest, aggregate and process streamed traffic data, enhanced with dynamic vehicle-level flow and urban event information, to produce regularly scheduled forecasts with high accuracy. In this work, we extend Foresight with several novel enhancements, into a new system which we term Foresight Plus. New features include an efficient method for extending the forecasting scale, enabling predictions further into the future. We also augment the inference architecture with a new, fully serverless design which offers a more cost-effective solution and which seamlessly handles sporadic inference workloads over multiple forecasting scales. We observe that Graph Neural Network (GNN) forecasting models are robust to extensions of the forecasting scale, achieving consistent performance up to 48 hours ahead. This is in contrast to the 1 hour forecasting periods popularly considered in this context. Further, our serverless inference solution is shown to be more cost-effective than provisioned alternatives in corresponding use-cases. We identify the optimal memory configuration of serverless resources to achieve an attractive cost-to-performance ratio.

The rise of machine learning and cloud technologies has led to a remarkable influx of data within modern cyber-physical systems. However, extracting meaningful information from this data has become a significant challenge due to its volume and complexity. Timed pattern matching has emerged as a powerful specification-based runtime verification and temporal data analysis technique to address this challenge. In this paper, we provide a comprehensive tutorial on timed pattern matching that ranges from the underlying algebra and pattern specification languages to performance analyses and practical case studies. Analogous to textual pattern matching, timed pattern matching is the task of finding all time periods within temporal behaviors of cyber-physical systems that match a predefined pattern. Originally we introduced and solved several variants of the problem using the name of match sets, which has evolved into the concept of timed relations over the past decade. Here we first formalize and present the algebra of timed relations as a standalone mathematical tool to solve the pattern matching problem of timed pattern specifications. In particular, we show how to use the algebra of timed relations to solve the pattern matching problem for timed regular expressions and metric compass logic in a unified manner. We experimentally demonstrate that our timed pattern matching approach performs and scales well in practice. We further provide in-depth insights into the similarities and fundamental differences between monitoring and matching problems as well as regular expressions and temporal logic formulas. Finally, we illustrate the practical application of timed pattern matching through two case studies, which show how to extract structured information from temporal datasets obtained via simulations or real-world observations. These results and examples show that timed pattern matching is a rigorous and efficient technique in developing and analyzing cyber-physical systems.

Building a real-time, cost-effective, spatio-temporal forecasting system is a challenging problem with many practical applications such as traffic and road network management. Most forecasting research focuses on average prediction quality, with less attention paid to building practical pipelines and achieving timely and accurate forecasts when the network is under heavy load. Additionally, transport authorities need to leverage dynamic data sources (e.g., scheduled roadworks) and vehicle-level flow data, while also supporting ad-hoc inference workloads at low cost. The cloud-based system Foresight, developed in collaboration with Transport for the West Midlands (TfWM), is able to ingest, aggregate and process streamed traffic data, as well as dynamic urban events/flow data to produce regularly scheduled forecasts with high accuracy. In this work, we extend our system with several novel enhancements. First, we present an efficient method for extending the forecasting scale, enabling transport managers to predict traffic patterns further into the future than existing methods. In addition, we augment the existing inference architecture with a new, fully serverless design. This offers a more cost-effective inference solution, which seamlessly handles sporadic inference workloads over multiple forecasting models. We observe that Graph Neural Network (GNN) forecasting models are robust to extensions of the forecasting scale, achieving consistent (and sometimes even improved) performance up to 24 hours ahead. This is in contrast to the 1 hour forecasting horizons popularly considered in the literature. Further, our serverless inference solution is shown to be significantly more cost-effective than provisioned alternatives in appropriate use-cases. We identify the optimal memory configuration of serverless resources to achieve an attractive cost-to-performance ratio.

Application of realism enhancement methods, particularly in real‐time and resource‐constrained settings, has been frustrated by the expense of existing methods. These achieve high quality results only at the cost of long runtimes and high bandwidth, memory, and power requirements. We present an efficient alternative: a high‐performance, generative shader‐based approach that adapts machine learning techniques to real‐time applications, even in resource‐constrained settings such as embedded and mobile GPUs. The proposed learnable shader pipeline comprises differentiable functions that can be trained in an end‐to‐end manner using an adversarial objective, allowing for faithful reproduction of the appearance of a target image set without manual tuning. The shader pipeline is optimized for highly efficient execution on the target device, providing temporally stable, faster‐than‐real time results with quality competitive with many neural network‐based methods.

Graph Convolutional Networks (GCNs) are extensively utilized for deep learning on graphs. The large data sizes of graphs and their vertex features make scalable training algorithms and distributed memory systems necessary. Since the convolution operation on graphs induces irregular memory access patterns, designing a memory- and communication-efficient parallel algorithm for GCN training poses unique challenges. We propose a highly parallel training algorithm that scales to large processor counts. In our solution, the large adjacency and vertex-feature matrices are partitioned among processors. We exploit the vertex-partitioning of the graph to use non-blocking point-to-point communication operations between processors for better scalability. To further minimize the parallelization overheads, we introduce a sparse matrix partitioning scheme based on a hypergraph partitioning model for full-batch training. We also propose a novel stochastic hypergraph model to encode the expected communication volume in mini-batch training. We show the merits of the hypergraph model, previously unexplored for GCN training, over the standard graph partitioning model which does not accurately encode the communication costs. Experiments performed on real-world graph datasets demonstrate that the proposed algorithms achieve considerable speedups over alternative solutions. The optimizations achieved on communication costs become even more pronounced at high scalability with many processors. The performance benefits are preserved in deeper GCNs having more layers as well as on billion-scale graphs.

Current approaches for modeling propagation in networks (e.g., of diseases, computer viruses, rumors) cannot adequately capture temporal properties such as order/duration of evolving connections or dynamic likelihoods of propagation along connections. Temporal models on evolving networks are crucial in applications that need to analyze dynamic spread. For example, a disease spreading virus has varying transmissibility based on interactions between individuals occurring with different frequency, proximity, and venue population density. Similarly, propagation of information having a limited active period, such as rumors, depends on the temporal dynamics of social interactions. To capture such behaviors, we first develop the Temporal Independent Cascade (T-IC) model with a spread function that efficiently utilizes a hypergraph-based sampling strategy and dynamic propagation probabilities. We prove this function to be submodular, with guarantees of approximation quality. This enables scalable analysis on highly granular temporal networks where other models struggle, such as when the spread across connections exhibits arbitrary temporally evolving patterns. We then introduce the notion of ‘reverse spread’ using the proposed T-IC processes, and develop novel solutions to identify both sentinel/detector nodes and highly susceptible nodes. Extensive analysis on real-world datasets shows that the proposed approach significantly outperforms the alternatives in modeling both if and how spread occurs, by considering evolving network topology alongside granular contact/interaction information. Our approach has numerous applications, such as virus/rumor/influence tracking. Utilizing T-IC, we explore vital challenges of monitoring the impact of various intervention strategies over real spatio-temporal contact networks where we show our approach to be highly effective.

In real-time rendering, a 3D scene is modelled with meshes of triangles that the GPU projects to the screen. They are discretized by sampling each triangle at regular space intervals to generate fragments which are then added texture and lighting effects by a shader program. Realistic scenes require detailed geometric models, complex shaders, high-resolution displays and high screen refreshing rates, which all come at a great compute time and energy cost. This cost is often dominated by the fragment shader, which runs for each sampled fragment. Conventional GPUs sample the triangles once per pixel; however, there are many screen regions containing low variation that produce identical fragments and could be sampled at lower than pixel-rate with no loss in quality. Additionally, as temporal frame coherence makes consecutive frames very similar, such variations are usually maintained from frame to frame. This work proposes Dynamic Sampling Rate (DSR), a novel hardware mechanism to reduce redundancy and improve the energy efficiency in graphics applications. DSR analyzes the spatial frequencies of the scene once it has been rendered. Then, it leverages the temporal coherence in consecutive frames to decide, for each region of the screen, the lowest sampling rate to employ in the next frame that maintains image quality. We evaluate the performance of a state-of-the-art mobile GPU architecture extended with DSR for a wide variety of applications. Experimental results show that DSR is able to remove most of the redundancy inherent in the color computations at fragment granularity, which brings average speedups of 1.68x and energy savings of 40%.

Contract-based design is a promising methodology for taming the complexity of developing sophisticated systems. A formal contract distinguishes between assumptions , which are constraints that the designer of a component puts on the environments in which the component can be used safely, and guarantees , which are promises that the designer asks from the team that implements the component. A theory of formal contracts can be formalized as an interface theory , which supports the composition and refinement of both assumptions and guarantees. Although there is a rich landscape of contract-based design methods that address functional and extra-functional properties, we present the first interface theory that is designed for ensuring system-wide security properties. Our framework provides a refinement relation and a composition operation that support both incremental design and independent implementability. We develop our theory for both stateless and stateful interfaces. We illustrate the applicability of our framework with an example inspired from the automotive domain.

We study the problem of specifying sequential information-flow properties of systems. Information-flow properties are hyperproperties, as they compare different traces of a system. Sequential information-flow properties can express changes, over time, in the information-flow constraints. For example, information-flow constraints during an initialization phase of a system may be different from information-flow constraints that are required during the operation phase. We formalize several variants of interpreting sequential information-flow constraints, which arise from different assumptions about what can be observed of the system. For this purpose, we introduce a first-order logic, called Hypertrace Logic, with both trace and time quantifiers for specifying linear-time hyperproperties. We prove that HyperLTL, which corresponds to a fragment of Hypertrace Logic with restricted quantifier prefixes, cannot specify the majority of the studied variants of sequential information flow, including all variants in which the transition between sequential phases (such as initialization and operation) happens asynchronously. Our results rely on new equivalences between sets of traces that cannot be distinguished by certain classes of formulas from Hypertrace Logic. This presents a new approach to proving inexpressiveness results for HyperLTL.

Timed pattern matching consists in finding occurrences of a timed regular expression in a timed word. This problem has been addressed using several techniques, its solutions are implemented in tools (quite efficient in practice), and used, for example in log analysis and runtime verification. In this article, we explore computational complexity of timed pattern matching, and prove P, NP and PSPACE bounds, depending on connectives used in expressions and other details. We conclude with a couple of open questions.

Modern cyber-physical systems (CPS) and the Internet of things (IoT) are data factories generating, measuring and recording huge amounts of time series. The useful information in time series is usually present in the form of sequential patterns. We propose shape expressions as a declarative language for specification and extraction of rich temporal patterns from possibly noisy data. Shape expressions are regular expressions with arbitrary (linear, exponential, sinusoidal, etc.) shapes with parameters as atomic predicates and additional constraints on these parameters. We associate with shape expressions novel noisy semantics that combines regular expression matching semantics with statistical regression. We study essential properties of the language and propose an efficient heuristic for approximate matching of shape expressions. We demonstrate the applicability of this technique on two case studies from the health and the avionics domains.

The most common task of GPUs is to render images in real time. When rendering a 3D scene, a key step is to determine which parts of every object are visible in the final image. There are different approaches to solve the visibility problem, the Z-Test being the most common. A main factor that significantly penalizes the energy efficiency of a GPU, especially in the mobile arena, is the so-called overdraw , which happens when a portion of an object is shaded and rendered but finally occluded by another object. This useless work results in a waste of energy; however, a conventional Z-Test only avoids a fraction of it. In this article we present a novel microarchitectural technique, the Omega-Test, to drastically reduce the overdraw on a Tile-Based Rendering (TBR) architecture. Graphics applications have a great degree of inter-frame coherence, which makes the output of a frame very similar to the previous one. The proposed approach leverages the frame-to-frame coherence by using the resulting information of the Z-Test for a tile (a buffer containing all the calculated pixel depths for a tile), which is discarded by nowadays GPUs, to predict the visibility of the same tile in the next frame. As a result, the Omega-Test early identifies occluded parts of the scene and avoids the rendering of non-visible surfaces eliminating costly computations and off-chip memory accesses. Our experimental evaluation shows average EDP savings in the overall GPU/Memory system of 26.4 percent and an average speedup of 16.3 percent for the evaluated benchmarks.

Current safety standards for automated driving recommend the development of a safety case. This case aims to justify and critically evaluate, by means of an explicit argument and evidence, how the safety claims concerning the intended functionality of an automated driving feature are supported. However, little guidance exists on how such an argument could be developed. In this paper, the MISRA consortium proposes a state machine on which an argument concerning the safety of the intended functionality could be structured. By systematically covering the activation status of the automated driving feature within and outside the operational design domain, this state machine helps in exploring the conditions, and asserting the corresponding safety claims, under which hazardous events could be caused by the intended functionality. MISRA uses a Traffic Jam Drive feature to illustrate the application of this approach.

Magnetic resonance imaging (MRI) can generate multimodal scans with complementary contrast information, capturing various anatomical or functional properties of organs of interest. But whilst the acquisition of multiple modalities is favourable in clinical and research settings, it is hindered by a range of practical factors that include cost and imaging artefacts. We propose XmoNet, a deep-learning architecture based on fully convolutional networks (FCNs) that enables cross-modality MR image inference. This multiple branch architecture operates on various levels of image spatial resolutions, encoding rich feature hierarchies suited for this image generation task. We illustrate the utility of XmoNet in learning the mapping between heterogeneous T1- and T2-weighted MRI scans for accurate and realistic image synthesis in a preliminary analysis. Our findings support scaling the work to include larger samples and additional modalities.

Measurements of the open circuit voltage of Li-ion cells have been extensively used as a non- destructive characterisation tool. Another technique based on entropy change measurements has also been applied for this purpose. More recently, both techniques have been used to make qualitative statements about aging in Li-ion cells. One proposed cause of cell failure is point defect formation in the electrode materials. The steps in voltage profiles, and the peaks in entropy profiles are sensitive to order/disorder transitions arising from Li/vacancy configurations, which are affected by the host lattice structures. We compare the entropy change results, voltage profiles and incremental capacity (dQ/dV ) obtained from coin cells with spinel lithium manganese oxide (LMO) cathodes, Li 1+y Mn 2-y O 4 , where excess Li y was added in the range 0 ≤y≤ 0.2. A clear trend of entropy and dQ/dV peak amplitude decrease with excess Li amount was determined. The effect arises, in part, from the presence of pinned Li sites, which disturb the formation of the ordered phase. We modelled the voltage, dQ/dV and entropy results as a function of the interaction parameters and the excess Li amount, using a mean field approach. For a given pinning population, we demonstrated that the asymmetries observed in the dQ/dV peaks can be modelled by a single linear correction term. To replicate the observed peak separations, widths and magnitudes, we had to account for variation in the energy interaction parameters as a function of the excess Li amount, y. All Li-Li repulsion parameters in the model increased in value as the defect fraction, y, increased. Our paper shows how far a computational mean field approximation can replicate experimentally observed voltage, incremental capacity and entropy profiles in the presence of phase transitions.

Dialysis patients in the UK usually undergo routine monthly blood tests which support the medical team in assessing their ongoing condition. Based on these results, clinicians then advise the patients on appropriate changes to diet and/or medication to improve their health. Whilst the results of these blood tests can be made available to the patient via an online patient portal, their presentation is primarily numerical. While this style is applicable to medical professionals who are able to interpret such results, it is less accessible to patients, restricting their ability to readily engage with their own results. This research presents a collaborative design approach aimed to produce alternative ways of visualising blood test results that meet both the needs of clinicians and the patient with the aim of enabling patients to be more actively involved in managing their own condition.

Microwave imaging is an emerging breast cancer diagnostic technique, which aims at complementing already established methods like mammography, magnetic resonance imaging, and ultrasound. It offers two striking advantages: no-risk for the patient and potential low-cost for national health systems. So far, however, the prototypes developed for validation in labs and clinics used costly lab instruments such as a vector network analyzer (VNA). Moreover, the CPU time required by complex image reconstruction algorithms may not be compatible with the duration of a medical examination. In this paper, both these issues are tackled. Indeed, we present a prototype system based on low-cost and off-the-shelf microwave components, custom-made antennas, and a small form-factor processing system with an embedded field-programmable gate array for accelerating the execution of the imaging algorithm. We show that our low-cost system can compete with an expensive VNA in terms of accuracy, and it is more than 20x faster than a high-performance server at image reconstruction.

In this paper we describe how to use MIPSfpga, a soft-core MIPS processor, to teach undergraduate and masters-level computer architecture courses. The most recent release of MIPSfpga (version 2.0), consists of three packages: the MIPSfpga Getting Started Guide, MIPSfpga Labs, and MIPSfpga System on Chip. After giving an overview of these packages, we provide examples of how to integrate MIPSfpga into curricula by describing three teaching experiences that used the MIPSfpga packages: an undergraduate course at the University Complutense of Madrid, a course at the Technical University of Darmstadt, and several seminars held at various Russian research centers and universities. MIPSfpga enabled students to bridge the gaps between theoretical concepts, hands-on practice, and industrial cores by allowing them to explore, modify, and test the MIPS core and system with the support of commercial compilers and tools.