Search Results (328)

In this paper, we develop modified inertial hybrid projection methods to obtain common solutions of generalized mixed equilibrium problems involving monotone and Lipschitz continuous operators. The use of alternative half-space constructions reduces both projection steps and operator evaluations. Strong convergence is established under standard conditions. The proposed approach is further extended to common solutions of variational inequality problems, and numerical results are provided for illustration. Full article

Mapping mountain surfaces requires spatial units that represent both hillslope-scale structure and local within-slope terrain heterogeneity. Hydrological slope units provide limited representation of within-slope objects, whereas general object-based segmentation is sensitive to fragmentation and scale selection. We developed a homogeneous terrain unit extraction framework based on superpixel segmentation and multiscale region merging (SSM-HTU), in which initial slope units serve as local statistical references and within-slope terrain objects are generated through slope-unit-conditioned morphometric representation, superpixel initialization, distribution-sensitive region merging, and a nested partition hierarchy. The framework was applied to the 5136 km² Yuqu River Basin in southeastern Tibet. Of 971 expert-interpreted reference HTUs, 680 were reserved for independent geometric evaluation; 1329 historical landslides were additionally used for supplementary spatial association analysis across mapping-unit schemes. Relative to the eCognition Multiresolution Segmentation (MSS) baseline, SSM-HTU showed a slight decrease in Precision from 0.8432 to 0.8340, while Recall (directional reference-object coverage) increased from 0.7615 to 0.8011, area-weighted IoU from 0.6710 to 0.6945, and Boundary F1 at a 12.5 m tolerance from 0.5980 to 0.6810, indicating greater reference-object coverage, spatial overlap, and boundary correspondence without uniform improvement across all geometric metrics. Across four geomorphological zones, area-weighted IoU ranged from 0.671 to 0.724 and Boundary F1 from 0.651 to 0.709, with non-monotonic regional variation. Mapping-unit schemes also yielded factor-dependent spatially stratified associations, underscoring the importance of spatial support in downstream statistical analysis. SSM-HTU therefore provides an object-based mapping framework for representing local within-slope terrain heterogeneity within a hillslope-scale statistical context in deeply incised valleys. Full article

In an infinite-dimensional Hilbert space, we investigate a method for finding a solution of a variational inequality induced by an inverse strongly monotone set-valued mapping. Our results are established in the presence of summable and nonsummable computational errors. We show that our algorithm generates a good approximate solution, if the sequence of computational errors is bounded from above by a constant. Full article

The transition from single-device characterization to array-level simulation remains a critical challenge in the development of three-terminal synaptic transistors for neuromorphic computing, as most existing simulation studies either extract parameters from a single representative device and apply them uniformly, or rely on weight-update strategies originally designed for two-terminal memristors. Here, we establish an experimentally calibrated behavioral simulation framework based on differential conductance-pair mapping (W = G⁺ − G⁻, where G⁺ and G⁻ denote the conductances of the positive and negative devices of each pair), integrating exponential long-term potentiation/depression (LTP/LTD) update rules with a posteriori screening mechanism (isValid) to systematically investigate how update polarity, step size, nonlinearity, and conductance boundaries regulate network computational efficiency. Through comprehensive simulation on the Neural Circuit Policies network, we demonstrate that the update direction must strictly align with the matrix’s role: the G⁻ channel requires unidirectional long-term depression inhibition, while the G⁺ channel can be frozen or bidirectionally updated. The optimal G⁻LTD and G⁺G⁻LTD strategies achieve accuracies of 0.9208 and 0.9481, respectively. Furthermore, we reveal a unique nonlinear gain effect under long-term depression > 0, where accuracy increases monotonically with nonlinearity level up to 0.9419. Device specification criteria are established: LTP-dominant updates favor large G_max, while LTD-dominant updates favor high G_min, with the G⁻LTD and G⁺G⁻LTD strategies showing accuracy fluctuations within ±0.005 across the tested boundary variations. Finally, the array-level implementation is validated through a functional-correctness check and device-parameter ablation experiments on a 23,715-weight array (47,430 differential conductance elements). This work provides an experimentally calibrated behavioral simulation platform and concrete algorithm-hardware co-design guidelines for future neuromorphic hardware, prioritizing synaptic devices with long-term depression > 0, a moderately elevated G_min, and asymmetric resource allocation toward LTD-side optimization. Full article

Mangrove deforestation is a global problem, but sediment-dominated estuaries can resist it through their own morphodynamic processes. This study reconstructs the spatiotemporal trajectory of the Ujung Pangkah estuary (1995–2025) to weigh natural progradation against anthropogenic pressure. We mapped the estuary in Google Earth Engine using Landsat archives and a Random Forest classifier with four spectral indices (NDVI, mNDWI, EVI, and MVI) and then independently validated all four epochs (Overall Accuracy 92.25–95.00%; Kappa 0.845–0.900). Error-adjusted change-detection analysis shows a non-monotonic trajectory: mangrove extent grew from 898.88 ha (1995) to a 2410.44 ha peak in 2015 and then contracted to 1816.87 ha by 2025 (error-adjusted: 1261.11 to 2626.03 to 2213.71 ha). Across the 30-year record, this is a statistically significant net expansion (z = 3.60, p < 0.001). Spatial attribution shows the 2015–2025 contraction comes mostly from landward anthropogenic conversion (78.1%), not seaward erosion (21.9%). A lagged correlation between a suspended sediment proxy and decadal net change (r = 0.92) offers quantitative support for continued sediment-driven coastal progradation. Ujung Pangkah’s resilience therefore coexists with a real, locatable anthropogenic pressure. Safeguarding this blue carbon ecosystem means targeting policy at the interior conversion zones already underway, alongside continued protection of the coastal frontier. Full article

Spacecraft radiation assessment requires expert interpretation of continuous environment-model outputs. We present a reflection-equivariant Mamdani fuzzy system for relative triage of modelled total ionising dose (TID) and solar-proton exposure. Radiation environment severity and solar-proton severity are derived from OMERE 5.9.5 runs of the AE9/AP9 (IRENE 1.57.004, mean mode) and Emission of Solar Protons (ESP, 90 per cent confidence) models, and, together with mission duration, are mapped through reflection-paired membership partitions and a 27-rule sum-based rule base to four triage categories. We prove reflection symmetry of the input and output partitions, permutation symmetry of the rule map, risk-reversal duality of the aggregated inference and centroid score, and reflection equivariance of a normalised output-support vector retained before defuzzification. The architecture is examined on nine reference mission scenarios and additional boundary cases using sensitivity, comparative-variant and cumulative-versus-duration-normalised analyses. The results show exact algebraic consistency with the imposed symmetry identities and transparent rule-level traceability, while also revealing the small local non-monotonicity of the centroid score and formulation sensitivity in the seven-year GLONASS-like scenario. Under the integrated-exposure formulation, scores range from 0.381 for the polar low-Earth-orbit scenario to 0.892 for the geostationary orbit (GEO). Because the same nine scenarios also define the frozen normalisation anchors, this range is a reference-set demonstration rather than an out-of-sample result. Evaluation to date comprises internal mathematical-consistency checks, comparison with an author-defined conservative heuristic and concordance with a seven-member expert panel blinded to the model output but rating the same scenario descriptions; the system has not been validated against ground-truth radiation-hardness outcomes such as mission anomaly records or component-qualification results. Cases for which the integrated and duration-normalised diagnostics disagree are flagged for separate engineering analysis. The system is a reference-benchmarked proof-of-concept pre-screening method and does not replace project-specific TID, total non-ionising dose (TNID), single-event-effect, shielding or component-qualification analysis. Full article

This paper introduces and analyzes an inertial projection iterative scheme for solving combined generalized general variational-like inequality problem (CGGVLIP), the zero problem associated with a $\gamma$-inverse strongly monotone mapping, and the common fixed-point problem of a finite family of relatively nonexpansive mappings in two-uniformly convex and uniformly smooth Banach spaces. Strong convergence of the proposed algorithm to a common solution is established. The efficiency of the method is illustrated through numerical examples and validated via detailed figures. An application to image denoising is also presented to demonstrate its practical utility. Full article

Vision-based traffic accident anticipation is critical for active vehicle safety systems, yet existing architectures frequently conflate performance gains with heavy parameter scaling optimized from scratch on compact domains. Consequently, during real-time inference, these frameworks suffer from severe prediction volatility and early triggering biases that induce dangerous control instability. To address these limitations, this paper shifts the research focus away from network modifications toward a highly controlled, strategy-driven training pipeline executed under a completely invariant spatial–temporal neural backbone. Our proposed paradigm establishes a robust framework through three decoupled milestones. First, an out-of-domain initialization strategy transferred generalized driving kinetics from a large-scale sequence domain (Mapillary) to serve as a stable temporal anchor. Second, a target-domain generative enrichment step injected synthetic nighttime scenes to decouple hazard features from low-light ambient noise. Third, progressive temporal supervision paradigm scaling targeted labels monotonically to align with continuous kinetic risk accumulation. Overall evaluations on the Car Crash Dataset (CCD) benchmark demonstrate that the fully integrated configuration (C4) pipeline achieves 69.89% in frame-level Mean Average Precision (mAP), which is an improvement of +22.81 percentage points over the baseline configuration. Continuous temporal measurements prove that our framework can adapt to tracking volatility, compressing Temporal Confidence Variance to 0.00328, and dropping the Prediction Instability Count to 0.66. While hyper-sensitive baselines report early raw latency averages driven by premature trigger noise, our model purposefully filters this early-frame variability to deliver a secure warning profile, achieving an absolute zero false alarm rate (FAR = 0.00%) across evaluated non-hazardous driving sequences, establishing the sequence-level trustworthiness required for practical autonomous deployment. Full article

Frost heave in seasonally frozen loess regions is fundamentally governed by pore water migration towards the freezing front driven by temperature gradients, forming ice lenses that damage engineered infrastructure. Because both freezing intensity and moisture availability evolve with climate, frost heave susceptibility is itself dynamic, yet existing assessments remain static and ignore future climate trajectories. This paper presents a physics-constrained machine learning framework that couples soil freezing characteristic curve (SFCC) prior knowledge with multi-source open data and CMIP6 climate projections to achieve dynamic frost heave susceptibility mapping for the Loess Plateau. Monotonicity constraints derived from the coupled phase-transition and cryosuction mechanisms described by the SFCC and the segregation potential theory are enforced during gradient-boosted tree training, ensuring that predictions respect the established relationships among freezing intensity, fine-grained content and ice segregation potential. An ordinal decomposition strategy is adopted to guarantee that the monotonicity constraint on each binary sub-model translates into monotonicity of the predicted ordinal susceptibility level. The best performer, physics-constrained XGBoost, reaches an overall accuracy of 88.7% and an AUC of 0.942 on a four-class susceptibility scheme. Independent validation against 156 field records and Sentinel-1 InSAR observations confirms that the model captures genuine frost heave patterns. Under SSP5-8.5, the area classified as high or very-high susceptibility contracts by approximately 38% by the 2080s owing to warming, while under SSP1-2.6 the reduction is only 12%, and transitional zones of moderate risk expand in both scenarios. These findings provide a temporally explicit and physically grounded basis for climate-adaptive infrastructure planning in cold loess regions. Full article

Agile outcome models commonly map Jira variables directly to a retrospective label and then explain the prediction through fragmented feature attributions; they rarely separate domain concepts, team clustering, unresolved work, and concept-label coupling. This study evaluates a domain-informed, concept-bottleneck-style explainable AI architecture for retrospective diagnosis of Agile Epic outcomes. A frozen Jira export of 10,000 unique issue-level records was linked to a pre-specified analytical cohort of 180 Epics across 14 teams. Six experts rated efficiency, effectiveness, sustainability, and contextual risk, while outcomes were recorded as Successful, Challenged, or Unsuccessful. Because the outcome labels and concept ratings were informed by the same Jira evidence, the models estimate consistency with an expert labelling procedure, rather than independent project success. Under five-fold group-aware cross-validation, the fixed-configuration flat LightGBM achieved macro-F1 = 0.864 ± 0.053 and the fixed-configuration HMXAI/CBM-style model achieved 0.843 ± 0.084. These descriptive primary scores are not a joint nested-model-selection comparison. The proposed method, therefore does, not demonstrate a performance improvement; its contribution is an inspectable diagnostic structure. Performance fell materially on the resolved-only subset (LightGBM macro-F1 = 0.645), and model-specific nested, leave-one-team-out, calibration, uncertainty, correlation, and intervention analyses further bound the claims. Concept interventions were not uniformly monotone, so the concept layer is domain-interpretable in form but not yet user-validated as actionable. The study contributes a transparent audit of when concept-level diagnosis can complement flat classification and when circularity, censoring, and shortcut learning restrict interpretation. Full article

We study a class of evolutionary variational inequalities in a Hilbert space that models the long-run balanced-growth equilibrium of a competitive economy with transaction costs, time-dependent production and infrastructure constraints, and exogenous price dynamics. The paper makes four contributions. First, we introduce a parametrized monotonicity functional ${\mu }_{\alpha }(t;F;u,v;p)$ and prove an exact equivalence theorem: the inequality ${\mu }_{\alpha }\ge \beta {∥u-v∥}^2$ holds if and only if the operator F is strongly monotone with the explicitly computed constant $m=\beta -\alpha (1+{∥p∥}_{\infty })$. This turns the growth parameter $\alpha$ and the price level into explicit terms of a single admissibility threshold and, for $\beta <\alpha (1+{∥p∥}_{\infty })$, produces a scale of conditions that covers operators which are not monotone, i.e., economies with a bounded degree of increasing returns. Second, we prove well-posedness: for every admissible initial state there is exactly one Lipschitz equilibrium trajectory $u^{*}(·)$, obtained through Moreau’s catching-up algorithm for the associated perturbed sweeping process, together with the explicit velocity bound $∥{\dot{u}}^{*}∥\le L_K+2C_F$. Third, we derive one comparison estimate from which global exponential stability, the convergence rate $∥u\left(t\right)-u^{*}\left(t\right)∥\le r e^{-mt}+L_p{m}^{-1}\sup ∥\Delta p∥+\epsilon m^{-1}$, and robustness with respect to perturbations of prices and of the operator all follow; we also show that, when the constraint sets stabilize, the trajectory converges to the stationary equilibrium of the limit problem. Fourth, we prove that strong monotonicity implies the c-covering property with $c=m$, so that the shock-absorbing capacity of the economy is governed by the same constant as the speed of convergence. Two examples—a two-resource system and an n-market network with nonlinear transaction costs—are worked out with a complete verification of every hypothesis and with explicit numerical constants. Full article

Atmospheric deposition in industrial basins of Central Asia is strongly influenced by local emissions and wintertime dispersion conditions. This study characterized snowpack at 63 sampling stations across Ust-Kamenogorsk, Kazakhstan, including operational background station 1, on 24–26 February 2025 after a 116-day accumulation period. Major ions were determined in a spatially distributed exploratory subset of 16 samples, and trace elements were measured in samples from all 63 stations by means of inductively coupled plasma mass spectrometry and optical emission spectrometry. Mean meltwater pH and total dissolved solids were 6.55 ± 0.34 and 37.3 ± 18.0 mg L⁻¹, respectively. Charge-balance errors for the 16 hydrochemical samples ranged from −0.3% to +0.7%. Using the contamination index based on exceedances of the current Kazakhstan water-quality thresholds, 48 stations had CI < 1, seven had CI = 1–3, and eight had CI > 3; the highest value (60.21) occurred at station 26. Principal component analysis showed that the first three components explained 53.6% of the variance and separated a broad mineral/industrial aerosol association from a Pb–Cd–Zn association consistent with non-ferrous metallurgy and mixed urban sources. Cadmium was therefore interpreted as the principal contributor to the MPC-normalized index at the most affected stations, rather than as the dominant component by absolute concentration. The dissolved fraction can be mobilized during spring melt, indicating a potential pathway to soils and receiving waters, although direct ecological or human-health risk was not quantified. Station-level point mapping and projection along the NW–SE axis showed localized multi-element maxima rather than a monotonic citywide gradient. Full article

We develop a sheaf-theoretic framework for shape analysis of covered shapes, where measure is introduced only after the underlying local-to-global topological structure has been established. Starting from a finite cubical complex together with a finite admissible cover, we build a finite topological model from the overlap structure of the cover and define a component sheaf on this model. The local data of the sheaf records the connected pieces visible on individual patches, while the maps between them describe how these pieces fit together across overlaps. The resulting degree-zero sheaf cohomology recovers the connected components of the full shape by the standard descent of locally constant functions. We show that the isomorphism class of this sheaf is an invariant of covered shapes, that it strictly refines ${\beta }_0$ for a fixed labeled cover and can distinguish some shapes with identical full Betti vectors, although it does not determine higher Betti numbers in general, and that it behaves functorially under symmetries preserving the cover. For dyadic covers, we distinguish the overlapping closed cover used by the sheaf from a paired half-open measurable partition, investigate measurable refinements, introduce monotone notions of local complexity under corrected refinement hypotheses, construct canonical sheaf-induced measures on both the index set and the ambient domain, and establish convergence and localization results for regular closed sets under pixel refinement. The sheaf-theoretic axiomatization therefore offers two complementary benefits: topologically, it provides a finite-space and cohomological invariant of covered shapes; measure-theoretically, it gives a principled hierarchy of quantitative summaries built only after the local gluing structure has been preserved. Full article

Graphite ore grade sorting is a key step toward intelligent mineral processing; however, it faces three representational contradictions: ambiguous classification posteriors at grade boundaries, asymmetric multi-scale feature interaction, and the mismatch between class-agnostic self-distillation assignment and sample-level difficulty. Targeting these, this paper adopts D-FINE as the baseline and introduces three decoupled improvements at its decoder, encoder, and criterion layers. (1) Boundary-Grade-aware Distribution Refinement (BG-FDR) online identifies boundary samples via the Top-2 classification score gap and modulates regression-distribution refinement, yielding +2.69 percentage points in mAP@0.5 with zero additional trainable parameters. (2) Bidirectional Feature Pyramid with Global–Local Spatial Attention (BiFPN-GLSA) builds a learnable weighted bidirectional multi-scale fusion path. (3) Difficulty-Aware Decoupled Distillation with Wise-Inner-Shape-IoU (DADD+Wise-IoU) imposes class- and sample-level difficulty-aware constraints. In the integrated full model, this increases Precision from 66.21% to 71.43% (+5.22 pp), F1 from 73.57% to 77.57%, and mean IoU from 97.81% to 98.35%, while false positives drop by 19.6%; the only parameter overhead (+3.84M) comes from BiFPN-GLSA, with BG-FDR and DADD adding effectively no network weights. Ablation on a self-built 3800-image dataset reveals a non-monotonic AP–Precision relationship: the mAP-optimal configuration (BG-FDR+BiFPN-GLSA, 94.17%) and the Precision-optimal one (DADD+Wise-IoU, 77.54%) do not coincide, providing a quantitative basis for objective-driven module selection in industrial sorting. Full article

In this work, scanning microwave microscopy (SMM) is used to monitor the evolution of the local microwave response, in terms of variations in the S₁₁ input reflection coefficient, in two III–V heterostructures relevant to mid-infrared photonics, namely a 150 nm thick heavily doped InGaAs layer on InP and HfO₂ (35 nm)/InGaAs (150 nm)/InP, subjected to Ga⁺ FIB milling over a broad dose range. By correlating raw, uncalibrated two-dimensional S₁₁ maps with atomic force microscopy (AFM) and Raman spectroscopy, we observe a dose-dependent evolution from implantation-dominated behavior to progressive amorphization, layer thinning and surface roughening. In the uncapped InGaAs/InP system, the SMM response varies monotonically with ion dose, consistent with progressive FIB-induced modification of the exposed InGaAs layer and, at larger milling depths, of the underlying InP substrate. In the HfO₂-capped structure, the microwave response is more complex: the oxide initially acts as a partial buffer against ion penetration, delaying damage transfer, but this effect progressively weakens as the cap is thinned and structurally degraded. The resulting S₁₁ contrast may reflect the combined effects of FIB-induced structural modifications, layer removal, local material composition, and surface morphology. Overall, the combined dataset indicates that SMM is a potentially highly sensitive probe of FIB-induced nanoscale modifications in the local microwave response, especially at low doses, provided that suitable on-chip calibration and de-embedding structures are available. Full article