Response timing measures play a crucial role in the assessment of automated driving systems (ADS) in collision avoidance scenarios, including but not limited to establishing human benchmarks and comparing ADS to human driver response performance. For example, measuring the response time (of a human driver or ADS) to a conflict requires the determination of a stimulus onset and a response onset. In existing studies, response onset relies on manual annotation or vehicle control signals such as accelerator and brake pedal movements. These methods are not applicable when analyzing large scale data where vehicle control signals are not available. This holds in particular for the rapidly expanding sets of ADS log data where the behavior of surrounding road users is observed via onboard sensors. To advance evaluation techniques for ADS and enable measuring response timing when vehicle control signals are not available, we developed a simple and efficient algorithm, based on a piecewise linear acceleration model, to automatically estimate brake onset that can be applied to any type of driving data that includes vehicle longitudinal time series data. We also proposed a manual annotation method to identify brake onset and used it as ground truth for validation. R^2 was used as a confidence metric to measure the accuracy of the algorithm, and its classification performance was analyzed using naturalistic collision avoidance data of both ADS and humans, where our method was validated against human manual annotation. Although our algorithm is subject to certain limitations, it is efficient, generalizable, applicable to any road user and scenario types, and is highly configurable.

Multimodal machine learning integrating histopathology and molecular data shows promise for cancer prognostication. We systematically reviewed studies combining whole slide images (WSIs) and high-throughput omics to predict overall survival. Searches of EMBASE, PubMed, and Cochrane CENTRAL (12/08/2024), plus citation screening, identified eligible studies. Data extraction used CHARMS; bias was assessed with PROBAST+AI; synthesis followed SWiM and PRISMA 2020. Protocol: PROSPERO (CRD42024594745). Forty-eight studies (all since 2017) across 19 cancer types met criteria; all used The Cancer Genome Atlas. Approaches included regularised Cox regression (n=4), classical ML (n=13), and deep learning (n=31). Reported c-indices ranged 0.550-0.857; multimodal models typically outperformed unimodal ones. However, all studies showed unclear/high bias, limited external validation, and little focus on clinical utility. Multimodal WSI-omics survival prediction is a fast-growing field with promising results but needs improved methodological rigor, broader datasets, and clinical evaluation. Funded by NPIC, Leeds Teaching Hospitals NHS Trust, UK (Project 104687), supported by UKRI Industrial Strategy Challenge Fund.

By upgrading an existing deep learning model with the knowledge provided by one of the oldest sets of grayscale satellite imagery, known as CORONA, we improved the AI model's attitude towards the automatic identification of archaeological sites in an envir onment which has been completely transformed in the last five decades, including the complete destruction of many of those same sites. The initial Bing - based convolutional network model was re - trained using CORONA satellite imagery for the district of Abu Ghraib, west of Baghdad, central Mesopotamian floodplain. The results were twofold and surprising. First, the detection precision obtained on the area of interest increased sensibly: in particular, the Intersection - over - Union (IoU) values, at the image segmentation level, surpassed 85%, while the general accuracy in detecting archeological sites reached 90%. Second, our re - trained model allowed the identification of four new sites of archaeological interest (confirmed through field verification), previously not identified by archaeologists with traditional techniques. This has confirmed the efficacy of using AI techniques and the CORONA imagery from the 1960s to discover archaeological sites currently no longer visible, a concrete breakthrough with significant consequences for the study of landscapes with vanishing archaeological evidence induced by anthropization.

Operations in disaster response, search \& rescue, and military missions that involve multiple agents demand automated processes to support the planning of the courses of action (COA). Moreover, traverse-affecting changes in the environment (rain, snow, blockades, etc.) may impact the expected performance of a COA, making it desirable to have a pool of COAs that are diverse in task distributions across agents. Further, variations in agent capabilities, which could be human crews and/or autonomous systems, present practical opportunities and computational challenges to the planning process. This paper presents a new theoretical formulation and computational framework to generate such diverse pools of COAs for operations with soft variations in agent-task compatibility. Key to the problem formulation is a graph abstraction of the task space and the pool of COAs itself to quantify its diversity. Formulating the COAs as a centralized multi-robot task allocation problem, a genetic algorithm is used for (order-ignoring) allocations of tasks to each agent that jointly maximize diversity within the COA pool and overall compatibility of the agent-task mappings. A graph neural network is trained using a policy gradient approach to then perform single agent task sequencing in each COA, which maximizes completion rates adaptive to task features. Our tests of the COA generation process in a simulated environment demonstrate significant performance gain over a random walk baseline, small optimality gap in task sequencing, and execution time of about 50 minutes to plan up to 20 COAs for 5 agent/100 task operations.

Deep neural networks are powerful tools for solving nonlinear problems in science and engineering, but training highly accurate models becomes challenging as problem complexity increases. Non-convex optimization and numerous hyperparameters to tune make performance improvement difficult, and traditional approaches often prioritize minimizing mean squared error (MSE) while overlooking $L^{\infty}$ error, which is the critical focus in many applications. To address these challenges, we present a progressive framework for training and tuning high-precision neural networks (HiPreNets). Our approach refines a previously explored staged training technique for neural networks that improves an existing fully connected neural network by sequentially learning its prediction residuals using additional networks, leading to improved overall accuracy. We discuss how to take advantage of the structure of the residuals to guide the choice of loss function, number of parameters to use, and ways to introduce adaptive data sampling techniques. We validate our framework's effectiveness through several benchmark problems.

In this very personal workography, I relate my 40-year experiences as a researcher and educator in and around Artificial Intelligence (AI), more specifically Natural Language Processing. I describe how curiosity, and the circumstances of the day, led me to work in both industry and academia, and in various countries, including The Netherlands (Amsterdam, Eindhoven, and Utrecht), the USA (Stanford), England (Brighton), Scotland (Aberdeen), and China (Beijing and Harbin). People and anecdotes play a large role in my story; the history of AI forms its backdrop. I focus on things that might be of interest to (even) younger colleagues, given the choices they face in their own work and life at a time when AI is finally emerging from the shadows.

Large Language Models (LLMs) are frequently utilized as sources of knowledge for question-answering. While it is known that LLMs may lack access to real-time data or newer data produced after the model's cutoff date, it is less clear how their knowledge spans across historical information. In this study, we assess the breadth of LLMs' knowledge using financial data of U.S. publicly traded companies by evaluating more than 197k questions and comparing model responses to factual data. We further explore the impact of company characteristics, such as size, retail investment, institutional attention, and readability of financial filings, on the accuracy of knowledge represented in LLMs. Our results reveal that LLMs are less informed about past financial performance, but they display a stronger awareness of larger companies and more recent information. Interestingly, at the same time, our analysis also reveals that LLMs are more likely to hallucinate for larger companies, especially for data from more recent years. The code, prompts, and model outputs are available on GitHub.

Continuous-time deep learning models, such as neural ordinary differential equations (ODEs), offer a promising framework for surrogate modeling of complex physical systems. A central challenge in training these models lies in learning expressive yet stable time-varying weights, particularly under computational constraints. This work investigates weight parameterization strategies that constrain the temporal evolution of weights to a low-dimensional subspace spanned by polynomial basis functions. We evaluate both monomial and Legendre polynomial bases within neural ODE and residual network (ResNet) architectures under discretize-then-optimize and optimize-then-discretize training paradigms. Experimental results across three high-dimensional benchmark problems show that Legendre parameterizations yield more stable training dynamics, reduce computational cost, and achieve accuracy comparable to or better than both monomial parameterizations and unconstrained weight models. These findings elucidate the role of basis choice in time-dependent weight parameterization and demonstrate that using orthogonal polynomial bases offers a favorable tradeoff between model expressivity and training efficiency.

Machine unlearning aims to remove the influence of specific samples from a trained model. A key challenge in this process is over-unlearning, where the model's performance on the remaining data significantly drops due to the change in the model's parameters. Existing unlearning algorithms depend on the remaining data to prevent this issue. As such, these methods are inapplicable in a more practical scenario, where only the unlearning samples are available (i.e., zero-shot unlearning). This paper presents a novel framework, ZS-P AG, to fill this gap. Our approach offers three key innovations: (1) we approximate the inaccessible remaining data by generating adversarial samples; (2) leveraging the generated samples, we pinpoint a specific subspace to perform the unlearning process, therefore preventing over-unlearning in the challenging zero-shot scenario; and (3) we consider the influence of the unlearning process on the remaining samples and design an influence-based pseudo-labeling strategy. As a result, our method further improves the model's performance after unlearning. The proposed method holds a theoretical guarantee, and experiments on various benchmarks validate the effectiveness and superiority of our proposed method over several baselines.

Small racing quadcopters carrying explosives, known as first-person-view drones or FPVs, have become the dominant weapon in the war in Ukraine. Now, some are fitted with solar cells so they can wait for extended periods to ambush targets, turning them into a new type of land mine. "The drone can sit by a road or choke point and when it acquires its target, it can then do a quick sprint to the target," says Robert Bunker at US consultancy firm C/O Futures. Drone ambushes, where the devices land beside a road or on a building and wait for a target, are already commonly carried out by both Russian and Ukrainian forces. But even with their engines turned off, their camera and radio communications drain the drones' battery, limiting waiting time to a few hours at best.