Value-aware AI should recognise human values and adapt to the value systems (value-based preferences) of different users. This requires operationalization of values, which can be prone to misspecification. The social nature of values demands their representation to adhere to multiple users while value systems are diverse, yet exhibit patterns among groups. In sequential decision making, efforts have been made towards personalization for different goals or values from demonstrations of diverse agents. However, these approaches demand manually designed features […]

Ver mais

When I started writing about AI in 2019, my first article was dedicated to what machine learning is and how it works. It explained the difference between weak and strong (or general) AI and mentioned that we have no idea how to create the latter. Well, in 2023, maybe we actually (finally!) do. I’m talking about multimodal learning. Don’t get me wrong – multimodal learning wasn’t a new concept already in 2019. Back then, TechRepublic called it the […]

Ver mais

arXiv:2601.20038v1 Announce Type: new Abstract: Kawarabayashi and Sidiropoulos [KS22] obtained an $O(log^2 n)$-approximation algorithm for Multicut in planar digraphs via a natural LP relaxation, which also establishes a corresponding upper bound on the multicommodity flow-cut gap. Their result is in contrast to a lower bound of $tilde{Omega}(n^{1/7})$ on the flow-cut gap for general digraphs due to Chuzhoy and Khanna [CK09]. We extend the algorithm and analysis in [KS22] to the node-weighted Multicut problem. Unlike in general digraphs, node-weighted […]

Ver mais

Organizations that have become more integrated into data science are changing from analytics to intelligence. Economic intelligence does not consist of building more models or dashboards; it is embedding a structured way of reasoning into decision systems. An econometrics-first approach provides a strong roadmap for designing those systems and to ensure that data-driven insights are communicated to action that is sound, and coherent. From Analytics Outputs to Decision Systems Conventional analytics pipelines usually have outputs of predictions or […]

Ver mais

arXiv:2604.13215v1 Announce Type: new Abstract: This work explores methods to identify energy system designs for infeasible control co-design optimization problems. Control co-design, or CCD, has been recognized as a powerful tool to maximize energy system capabilities through simultaneous determination of plant and controller parameters. However, due to the inherent nonlinearities, complexity, and conflicting criteria of energy systems, CCD optimization problems are susceptible to infeasibility and can lack potential solutions. While transforming the optimization problem by relaxing constraints has […]

Ver mais

arXiv:2603.05519v1 Announce Type: new Abstract: The rampant spread of fake news in the digital age poses serious risks to public trust and democratic institutions, underscoring the need for effective, transparent, and user-centered detection tools. Existing browser extensions often fall short due to opaque model behavior, limited explanatory support, and a lack of meaningful user engagement. This paper introduces Aletheia, a novel browser extension that leverages Retrieval-Augmented Generation (RAG) and Large Language Models (LLMs) to detect fake news and […]

Ver mais

arXiv:2603.20025v2 Announce Type: replace Abstract: We study adversarial learning when the target distribution factorizes according to a known Bayesian network. For interpolative divergences, including $(f,Gamma)$-divergences, we prove a new infimal subadditivity principle showing that, under suitable conditions, a global variational discrepancy is controlled by an average of family-level discrepancies aligned with the graph. In an additive regime, the surrogate is exact. This closes a theoretical gap in the literature; existing subadditivity results justify graph-informed adversarial learning for classical […]

Ver mais