The integration of predictive maintenance and cybersecurity represents a transformative advancement for small and medium-sized enterprises (SMEs) operating within the Industry 4.0 paradigm. Despite their economic importance, SMEs often face significant challenges in adopting advanced technologies due to resource constraints and knowledge gaps. The DETECTA 2.0 project addresses these hurdles by developing an innovative system that harmonizes real-time anomaly detection, sophisticated analytics, and predictive forecasting capabilities. The system employs a semi-supervised methodology, combining unsupervised anomaly detection with supervised learning techniques. This approach enables more agile and cost-effective development of AI detection systems, significantly reducing the time required for manual case review. At the core lies a Digital Twin interface, providing intuitive real-time visualizations of machine states and detected anomalies. Leveraging cutting-edge AI engines, the system intelligently categorizes anomalies based on observed patterns, differentiating between technical errors and potential cybersecurity incidents. This discernment is fortified by detailed analytics, including certainty levels that enhance alert reliability and minimize false positives. The predictive engine uses advanced time series algorithms like N-HiTS to forecast future machine utilization trends. This proactive approach optimizes maintenance planning, enhances cybersecurity measures, and minimizes unplanned downtimes despite variable production processes. With its modular architecture enabling seamless integration across industrial setups and low implementation costs, DETECTA 2.0 presents an attractive solution for SMEs to strengthen their predictive maintenance and cybersecurity strategies.

We propose a novel approach for training large language models (LLMs) to adhere to objectives defined within a latent embedding space. Our method leverages reinforcement learning (RL), treating a pre-trained LLM as an environment. Our embedding-aligned guided language (EAGLE) agent is trained to iteratively steer the LLM's generation towards optimal regions of the latent embedding space, w.r.t. some predefined criterion. We demonstrate the effectiveness of the EAGLE agent using the MovieLens 25M dataset to surface content gaps that satisfy latent user demand. We also demonstrate the benefit of using an optimal design of a state-dependent action set to improve EAGLE's efficiency. Our work paves the way for controlled and grounded text generation using LLMs, ensuring consistency with domain-specific knowledge and data representations.

Near-real time estimation of damage to buildings and infrastructure, referred to as damage nowcasting in this study, is crucial for empowering emergency responders to make informed decisions regarding evacuation orders and infrastructure repair priorities during disaster response and recovery. Here, we introduce FloodDamageCast, a machine learning (ML) framework tailored for property flood damage nowcasting. The framework leverages heterogeneous data to predict residential flood damage at a resolution of 500 meters by 500 meters within Harris County, Texas, during the 2017 Hurricane Harvey. To deal with data imbalance, FloodDamageCast incorporates a generative adversarial networks-based data augmentation coupled with an efficient machine learning model. The results demonstrate the framework's ability to identify high-damage spatial areas that would be overlooked by baseline models. Insights gleaned from flood damage nowcasting can assist emergency responders to more efficiently identify repair needs, allocate resources, and streamline on-the-ground inspections, thereby saving both time and effort. Keywords: Flood damage nowcasting Data augmentation Generative adversarial network Light gradient-boosting machine Imbalance learning 1 Introduction Flood hazards wreak havoc on urban areas, resulting in both physical destruction and loss of life in densely populated regions. In the United States alone, annual insurance claims have hovered around $1 billion per year over the past four decades [1]. This financial burden is expected to persist and potentially worsen due to the escalating frequency and intensity of flood events resulting from climate change [2, 3]. Rapid damage assessment of flooded areas is essential for swift response and recovery of affected communities. Emergency responders and public officials rely primarily on visual inspection to evaluate flood damage, incurring significantly delaying the recovery process. Expediting the flood damage assessment process is instrumental to accelerating post-disaster recovery efforts and bolstering community resilience against flood hazards, Currently, the main approach for estimating flood damage is based on specifying inundation depths then utilizing historical flood depth damage curves [4, 5]. The applicability of this approach for flood damage nowcasting, however, would be limited due to significant computation effort needed to model inundation depths using hydrological models based on the principles of hydrodynamics [6, 7, 8, 9].

This study proposes two new dynamic assignment algorithms to match refugees and asylum seekers to geographic localities within a host country. The first, currently implemented in a multi-year randomized control trial in Switzerland, seeks to maximize the average predicted employment level (or any measured outcome of interest) of refugees through a minimum-discord online assignment algorithm. The performance of this algorithm is tested on real refugee resettlement data from both the US and Switzerland, where we find that it is able to achieve near-optimal expected employment compared to the hindsight-optimal solution, and is able to improve upon the status quo procedure by 40-50%. However, pure outcome maximization can result in a periodically imbalanced allocation to the localities over time, leading to implementation difficulties and an undesirable workflow for resettlement resources and agents. To address these problems, the second algorithm balances the goal of improving refugee outcomes with the desire for an even allocation over time. We find that this algorithm can achieve near-perfect balance over time with only a small loss in expected employment compared to the employment-maximizing algorithm. In addition, the allocation balancing algorithm offers a number of ancillary benefits compared to pure outcome maximization, including robustness to unknown arrival flows and greater exploration.

In the next few years, applications of Generative AI are expected to revolutionize a number of different areas, ranging from science & medicine to education. The potential for these seismic changes has triggered a lively debate about potential risks and resulted in calls for tighter regulation, in particular from some of the major tech companies who are leading in AI development. This regulation is likely to put at risk the budding field of open-source Generative AI. We argue for the responsible open sourcing of generative AI models in the near and medium term. To set the stage, we first introduce an AI openness taxonomy system and apply it to 40 current large language models. We then outline differential benefits and risks of open versus closed source AI and present potential risk mitigation, ranging from best practices to calls for technical and scientific contributions. We hope that this report will add a much needed missing voice to the current public discourse on near to mid-term AI safety and other societal impact.

Diffusion models (DMs) based adversarial purification (AP) has shown to be the most powerful alternative to adversarial training (AT). However, these methods neglect the fact that pre-trained diffusion models themselves are not robust to adversarial attacks as well. Additionally, the diffusion process can easily destroy semantic information and generate a high quality image but totally different from the original input image after the reverse process, leading to degraded standard accuracy. To overcome these issues, a natural idea is to harness adversarial training strategy to retrain or fine-tune the pre-trained diffusion model, which is computationally prohibitive. We propose a novel robust reverse process with adversarial guidance, which is independent of given pre-trained DMs and avoids retraining or fine-tuning the DMs. This robust guidance can not only ensure to generate purified examples retaining more semantic content but also mitigate the accuracy-robustness trade-off of DMs for the first time, which also provides DM-based AP an efficient adaptive ability to new attacks. Extensive experiments are conducted on CIFAR-10, CIFAR-100 and ImageNet to demonstrate that our method achieves the state-of-the-art results and exhibits generalization against different attacks.

Recent regulatory initiatives like the European AI Act and relevant voices in the Machine Learning (ML) community stress the need to describe datasets along several key dimensions for trustworthy AI, such as the provenance processes and social concerns. However, this information is typically presented as unstructured text in accompanying documentation, hampering their automated analysis and processing. In this work, we explore using large language models (LLM) and a set of prompting strategies to automatically extract these dimensions from documents and enrich the dataset description with them. Our approach could aid data publishers and practitioners in creating machine-readable documentation to improve the discoverability of their datasets, assess their compliance with current AI regulations, and improve the overall quality of ML models trained on them. In this paper, we evaluate the approach on 12 scientific dataset papers published in two scientific journals (Nature's Scientific Data and Elsevier's Data in Brief) using two different LLMs (GPT3.5 and Flan-UL2). Results show good accuracy with our prompt extraction strategies. Concrete results vary depending on the dimensions, but overall, GPT3.5 shows slightly better accuracy (81,21%) than FLAN-UL2 (69,13%) although it is more prone to hallucinations. We have released an open-source tool implementing our approach and a replication package, including the experiments' code and results, in an open-source repository.

Systems based on machine learning may exhibit discriminatory behavior based on sensitive characteristics such as gender, sex, religion, or race. In light of this, various notions of fairness and methods to quantify discrimination were proposed, leading to the development of numerous approaches for constructing fair predictors. At the same time, imposing fairness constraints may decrease the utility of the decision-maker, highlighting a tension between fairness and utility. This tension is also recognized in legal frameworks, for instance in the disparate impact doctrine of Title VII of the Civil Rights Act of 1964 -- in which specific attention is given to considerations of business necessity -- possibly allowing the usage of proxy variables associated with the sensitive attribute in case a high-enough utility cannot be achieved without them. In this work, we analyze the tension between fairness and accuracy from a causal lens for the first time. We introduce the notion of a path-specific excess loss (PSEL) that captures how much the predictor's loss increases when a causal fairness constraint is enforced. We then show that the total excess loss (TEL), defined as the difference between the loss of predictor fair along all causal pathways vs. an unconstrained predictor, can be decomposed into a sum of more local PSELs. At the same time, enforcing a causal constraint often reduces the disparity between demographic groups. Thus, we introduce a quantity that summarizes the fairness-utility trade-off, called the causal fairness/utility ratio, defined as the ratio of the reduction in discrimination vs. the excess loss from constraining a causal pathway. This quantity is suitable for comparing the fairness-utility trade-off across causal pathways. Finally, as our approach requires causally-constrained fair predictors, we introduce a new neural approach for causally-constrained fair learning.

The first patient living with Elon Musk's Neuralink'brain chip' implant wants the world to know how'amazing' and'rewarding' his clinical trial with the tech has been. Just four months ago, 30-year-old Noland Arbaugh went under the knife for the experimental surgery that would allow him to control computers with his mind. 'I'm really excited to keep going,' Arbaugh, who has been paralyzed from his neck's fourth vertebra down since college, said of his role in Neuralink's human trial. But while the tech company's brain-computer interface has allowed him to race his stepfather in Nintendo's Mario Kart, navigate a computer cursor and more -- all with just his thoughts alone -- technical hurdles still plague the brain chip's functioning. A report on Arbaugh's Neuralink trial said that approximately 85 percent of the chip's tendril-like connections to his brain have come loose, forcing Neuralink staff to retool the system on its software side, as FDA approves trials on a second patient.

The substance of the symposium addressed the challenges in creating synergistic human and AI-based autonomous systems-of-systems. Recent advances in generative AI techniques (e.g., LLMs) have exacerbated the growing concerns associated with AI, held by researchers and the public alike, such as the risk, trust, ethics, and safety to the users and to the public from the operations of autonomous machines/AI alone in open situations. These concerns present major hurdles in the development of verified and validated engineered systems involving bi-directional pathways across the human-machine barrier; in this context, bi-directionality means understanding the design and operational consequences that the human may have on machine agents and the effects that machine or AI agents may have on humans. Current discussions on human-AI/machine interactions are unresolved or fragmented, focusing either on the impact that AI or machines may have on human stakeholders (including the relevant human factor considerations) or potential ways of involving humans or machines in computational or physical interventions (e.g., data annotations, human-machine behavior interpretations, operator-machine interventions). We believe the challenges associated with human-AI/machine collaborative systems cannot be adequately addressed if the underlying challenges associated with bi-directionality are not fully identified and taken into consideration.