26838 research outputs found
Sort by
VRD: A multi-lingual translation and AI-assisted navigation experience for VR conference application
Get PDFThis paper presents an AI-assisted VR conference application with multilingual translation and navigation agent capabilities. A pilot study with 18 participants (11 females, 7 males) was conducted to assess the system’s usability. AI-assisted navigation worked smoothly, but the AI translation had issues that prevented the users from having a good experience, nonetheless, participants expressed positive attitudes toward the system, and future work will focus on achieving better user experience
Internet of multisensory, multimedia and musical things (Io3MT) environments: Requirements, use cases, and evaluation
Get PDFThe Internet of Multisensory, Multimedia, and Musical Things (Io3MT) can be understood as a transmission network that integrates, within a unified ecosystem, devices and data capable of engaging the five human senses (touch, hearing, vision, smell, and taste), multimedia content, and music information in an interchangeable and non-hierarchical manner, thereby providing globally accessible applications and services. This thesis introduces the first reference model of this domain, outlining its standard architecture, data types, communication requirements, and the tools suitable for its implementation. The objective is to establish guidelines that enable scientists, artists, designers, and industry practitioners to conceive and develop environments grounded in the principles of Io3MT. To validate the proposed framework, two proof-of-concept implementations were conducted, each exploring distinct dimensions of the paradigm. In these prototypes, experiments were carried out to evaluate network performance and to assess the resulting Quality of Experience (QoE). The experiments comprised a device prototype and an immersive environment designed for real-time artistic performance. The results confirmed the technical feasibility of the reference model while also demonstrating its aesthetic and expressive potential, thereby highlighting Io3MT’s capacity to sustain interactive, creative, and multisensory experiences
On the complexity of knapsack under explorable uncertainty: Hardness and algorithms
Get PDFIn the knapsack problem under explorable uncertainty, we are given a knapsack instance with uncertain item profits. Instead of having access to the precise profits, we are only given uncertainty intervals that are guaranteed to contain the corresponding profits. The actual item profit can be obtained via a query. The goal of the problem is to adaptively query item profits until the revealed information suffices to compute an optimal (or approximate) solution to the underlying knapsack instance. Since queries are costly, the objective is to minimize the number of queries. In the offline variant of this problem, we assume knowledge of the precise profits and the task is to compute a query set of minimum cardinality that a third party without access to the profits could use to identify an optimal (or approximate) knapsack solution. We show that this offline variant is complete for the second-level of the polynomial hierarchy, i.e., Σp2-complete, and cannot be approximated within a non-trivial factor unless Σp2 = ∆p2. Motivated by these strong hardness results, we consider a “resource-augmented” variant of the problem where the requirements on the query set computed by an algorithm are less strict than the requirements on the optimal solution we compare against. More precisely, a query set computed by the algorithm must reveal sufficient information to identify an approximate knapsack solution, while the optimal query set we compare against has to reveal sufficient information to identify an optimal solution. We show that this resource-augmented setting allows interesting non-trivial algorithmic results
ExcluIR: Exclusionary neural information retrieval
Get PDFExclusion is an important and universal linguistic skill that humans use to express what they do not want. There is little research on exclusionary retrieval, where users express what they do not want to be part of the results produced for their queries. We investigate the scenario of exclusionary retrieval in document retrieval for the first time. We present ExcluIR, a set of resources for exclusionary retrieval, consisting of an evaluation benchmark and a training set for helping retrieval models to comprehend exclusionary queries. The evaluation benchmark includes 3,452 high-quality exclusionary queries, each of which has been manually annotated. The training set contains 70,293 exclusionary queries, each paired with a positive document and a negative document. We conduct detailed experiments and analyses, obtaining three main observations: (i) existing retrieval models with different architectures struggle to comprehend exclusionary queries effectively; (ii) although integrating our training data can improve the performance of retrieval models on exclusionary retrieval, there still exists a gap compared to human performance; and (iii) generative retrieval models have a natural advantage in handling exclusionary queries
Is Ockham's razor losing its edge? New perspectives on the principle of model parsimony
Get PDFThe preference for simple explanations, known as the parsimony principle, has long guided the development of scientific theories, hypotheses, and models. Yet recent years have seen a number of successes in employing highly complex models for scientific inquiry (e.g., for 3D protein folding or climate forecasting). In this paper, we reexamine the parsimony principle in light of these scientific and technological advancements. We review recent developments, including the surprising benefits of modeling with more parameters than data, the increasing appreciation of the context-sensitivity of data and misspecification of scientific models, and the development of new modeling tools. By integrating these insights, we reassess the utility of parsimony as a proxy for desirable model traits, such as predictive accuracy, interpretability, effectiveness in guiding new research, and resource efficiency. We conclude that more complex models are sometimes essential for scientific progress, and discuss the ways in which parsimony and complexity can play complementary roles in scientific modeling practice
Conditional Direct Empirical Linkage Discovery for Solving Multi-Structured Problems
Get PDFMany state-of-the-art Genetic Algorithms (GAs) use information about variable dependencies to construct masks for variation operators and, in turn, improve their effectiveness and efficiency. In the black-box setting, the dependency structure model is not known and must be discovered as a part of the optimization process. The precision of this model may be decisive for the effectiveness of the GAs using it. This work considers the recently identified multi-structured problems that arise when two or more problems with a different structure (i.e., different variable dependencies) are combined in a non-linear manner. Such problems are hard to solve because, usually, it is not enough to know all the dependencies to solve them effectively. To do so, one must know which dependencies are a part of which substructure, i.e., the dependencies between dependencies. Finally, an optimizer must detect which substructure is valid for the solution at hand. Statistical Linkage Learning (SLL) was proposed to decompose multi-structure problems. However, SLL may report false dependencies, which can deteriorate the search. Therefore, we propose the Conditional Direct Empirical Linkage Discovery (cDLED) technique to decompose multi-structured problems. cDLED guarantees to report only true dependencies. Using cDLED, we propose detecting which problem substructure refers to the given solution. Using these two mechanisms, we propose an optimizer that is highly competitive with other state-of-the-art GAs. We consider single-objective optimization, but our propositions can also be useful in multi- and many-objective optimization. Additionally, we propose a more general formal representation of multi-structured problems
The algebraic degree of coupled oscillators
No full textApproximating periodic solutions to the coupled Duffing equations amounts to solving a system of polynomial equations. The number of complex solutions measures the algebraic complexity of this approximation problem. Using the theory of Khovanskii bases, we show that this number is given by the volume of a polytope. We also show how to compute all solutions using numerical nonlinear algebra
Complex speckle illumination for quantitative phase imaging
No full textSpeckle patterns generated by multimode fibers (MMFs) are often seen as a challenge in imaging, particularly when precise control of illumination is needed [1]. In this work, we focus on a different perspective: speckles from MMFs can be highly beneficial, offering a diversity of random illumination patterns. We turn this complex illumination into a powerful tool for robust and accurate non-interferometric quantitative phase imaging (QPI). QPI is a robust, label-free optical imaging technique used to study transparent or weakly scattering samples, such as biological cells, thin films, and microstructures [2]
MultiFIX: Interpretable Multimodal AI via Deep Learning and Genetic Programming
No full textMultiFIX is a multimodal feature engineering and fusion pipeline designed for interpretable AI. It combines:
Deep Learning (DL) for modality-specific feature extraction.
Genetic Programming (GP-GOMEA) for interpretable symbolic expressions.
Grad-CAM for visual explanations in image-based models.
MultiFIX supports tabular + image data and enforces compact, interpretable features per modality, followed by interpretable fusion models
Ransomware negotiation: Dynamics and privacy-preserving mechanism design
Get PDFRansomware attacks have become a pervasive and costly form of cybercrime, causing tens of millions of dollars in losses as organizations increasingly pay ransoms to mitigate operational disruptions and financial risks. While prior research has largely focused on proactive defenses, the post-infection negotiation dynamics between attackers and victims remains underexplored. This paper presents a formal analysis of attacker–victim interactions in modern ransomware incidents using a finite-horizon alternating-offers bargaining game model. Our analysis demonstrates how bargaining alters the optimal strategies of both parties. In practice, incomplete information—attackers lacking knowledge of victims’ data valuations and victims lacking knowledge of attackers’ reservation ransoms—can prolong negotiations and increase victims’ business interruption costs. To address this, we design a Bayesian incentive-compatible mechanism that facilitates rapid agreement on a fair ransom without requiring either party to disclose private valuations. We further implement this mechanism using secure two-party computation based on garbled circuits, thereby eliminating the need for trusted intermediaries and preserving the privacy of both parties throughout the negotiation. To the best of our knowledge, this is the first automated, privacy-preserving negotiation mechanism grounded in a formal analysis of ransomware negotiation dynamics