Abstract
We address a classification problem where data are not single-valued, but distributions. The objective is to identify Internet traffic re-direction. Each observation consists of a block of 10 measurements of round-trip-times (RTT) measured at each of a set of probes, and is represented by the corresponding empirical distribution. The proposed approach relies on a method for discriminant analysis of distributional data that uses fractional programming, and where distributions are represented by quantile functions, under specific assumptions. A linear discriminant function is defined, that allows obtaining a score for each unit, in the form of a quantile function. This is then used to classify the units in a priori groups, using the Mallows distance. Results show that proposed approach works well, allowing for the identification of the diverted traffic.

Abstract

Abstract
Inspecting reflective parts is challenging due to strong specular reflections that conceal small porosities and reduce defect visibility. This work presents a framework that combines augmented reality with a deep learning detector. An augmented reality headset is used to capture multi-view images under natural illumination, enabling the operator to adjust the viewpoint and obtain angles that reduce glare. The collected data form a 640x480 dataset used to train a yolov8 detection model, integrated into a Robot Operating System 2 architecture for realtime processing. Testing on an independent set of unseen parts yields a precision of 86.70%, a recall of 87.26%, and an F1score of 86.97%. Additional qualitative examples confirm that the model can identify low-contrast porosities despite reflective surfaces. The results demonstrate the feasibility of AR-assisted acquisition combined with deep learning for real-time inspection of machined aluminum components in a laboratory case study.

Abstract

Abstract
A central issue in price competition with positive network effects is the potential for small price changes to trigger abrupt chain reactions, leading to market tipping, winner-take-all scenarios, and zero-profit equilibria. We show that in a duopoly where consumers are not anonymous but partitioned into at least two groups, a simple group-based network structure can, by itself, generate downward-sloping demand and support profitable shared-market equilibria. These are subgame-perfect pure price equilibria in which both firms earn strictly positive profit. Triggering a bandwagon effect and tipping the market remains possible, but requires aggressive price deviations, or price shocks, that produce demand jumps. However, this is not always profitable, and the fear of bankruptcy can be sufficient to stabilize firms in equilibrium. The result relies on having one group with centripetal influence (stronger impact on peers) and another with centrifugal influence (stronger impact on outsiders). It requires no additional sources of heterogeneity or product differentiation. This mechanism shows that positive network effects - when group structured - can endogenously generate stability in price competition. The analysis reconciles the coexistence of local stability and the potential for tipping, offering a unified explanation of how markets with strong network effects can sustain both competition and profitability. We draw a parallel to Turing's reaction-diffusion patterns and reinterpret Becker's intuition that social influence can produce stable outcomes, even when demand may exhibit upward-sloping segments.

Abstract
Reliable localization without GPS is a prerequisite for autonomous unmanned aerial vehicles (UAVs) operating inside warehouses, where a lightweight monocular camera paired with an inertial measurement unit (IMU) and onboard computer are the minimal sensing and processing an onboard platform can carry. This paper presents a tightly coupled monocular point-feature visual-inertial odometry (VIO) system for that setting, realized on a GTSAM fixed-lag factor graph with inverse-depth landmarks, on-manifold IMU preintegration, and an online loop-closure pose graph. The system is developed as the initial estimation stage of an autonomous stock-management UAV under development for indoor logistics warehouses. The decisive design element is the bootstrap: the metric, gravity-aligned initialization of a monocular estimator is well conditioned only under a translation-rich trajectory, a condition the near-zero-baseline pickup and takeoff transient that opens every indoor flight violates. Building on the visual-inertial alignment of VINS-Mono, we harden this step with a pre-bundle-adjust conditioning gate and a continuous-window initialization that refines the whole bootstrap window inside the smoother instead of freezing a single seed. On all eleven EuRoC MAV sequences, indoor flight tests recorded onboard a micro air vehicle in an industrial hall and two instrumented rooms, one fixed configuration per operating environment converges on every sequence, including three that otherwise diverge by tens to thousands of meters, and, driven by the same feature stream as locally run VINS-Mono and PL-VINS baselines, attains the better pure-odometry accuracy on nine of the eleven, with ATE RMSE of 0.12–0.37 m on the Machine Hall, a margin a paired signed-rank test confirms against VINS-Mono and leaves unconfirmed against PL-VINS at this sample size. We identify the stock fixed-lag marginalization as the principal consistency limitation and outline First-Estimates-Jacobian marginalization as the route to a more consistent estimator, establishing a characterized point-only baseline on one public benchmark as the starting point for subsequent on-platform work.