Search for: All records

Artificial intelligence (AI) supported network traffic classification (NTC) has been developed lately for network measurement and quality-of-service (QoS) purposes. More recently, federated learning (FL) approach has been promoted for distributed NTC development due to its nature of unshared dataset for better privacy and confidentiality in raw networking data collection and sharing. However, network measurement still require invasive probes and constant traffic monitoring. In this paper, we propose a non-invasive network traffic estimation and user profiling mechanism by leveraging label inference of FL-based NTC. In specific, the proposed scheme only monitors weight differences in FL model updates from a targeting user and recovers its network application (APP) labels as well as a rough estimate on the traffic pattern. Assuming a slotted FL update mechanism, the proposed scheme further maps inferred labels from multiple slots to different profiling classes that depend on, e.g., QoS and APP categorization. Without loss of generality, user profiles are determined based on normalized productivity, entertainment, and casual usage scores derived from an existing commercial router and its backend server. A slot extension mechanism is further developed for more accurate profiling beyond raw traffic measurement. Evaluations conducted on seven popular APPs across three user profiles demonstrate that our approach can achieve accurate networking user profiling without invasive physical probes nor constant traffic monitoring. 

A critical use case of SLAM for mobile robots is to support localization during task-directed navigation. Current SLAM benchmarks overlook the importance of repeatability (precision) despite its impact on real-world deployments. TaskSLAM-Bench, a task-driven approach to SLAM benchmarking, addresses this gap. It employs precision as a key metric, accounts for SLAM’s mapping capabilities, and has easy-to-meet requirements. Simulated and real-world evaluation of SLAM methods provide insights into the navigation performance of modern visual and LiDAR SLAM solutions. The outcomes show that passive stereo SLAM precision may match that of 2D LiDAR SLAM in indoor environments. TaskSLAM-Bench complements existing benchmarks and offers richer assessment of SLAM performance in navigation-focused scenarios. Publicly available code permits in-situ SLAM testing in custom environments with properly equipped robots. 

This research-to-practice full paper describes a cohort-based undergraduate research program designed to improve STEM retention through structured mentoring and community building. Drawing on the Affinity Research Group (ARG) model, the program fosters faculty-student research collaboration and integrates faculty mentorship training, student-led peer mentoring, and structured interventions, such as research skills workshops and networking events. Each year, faculty from biology, chemistry, computer science, environmental science, and mathematics lead small-group research projects with recruited students who may participate for up to three years. Faculty and students receive ARG training to promote consistent mentoring practices. A credit-bearing, major-specific first-year orientation course supports recruitment and reinforces students’ scientific identity. Faculty also engage in professional development workshops to strengthen student-centered mentoring approaches. Data collection includes surveys, interviews, retention tracking, and weekly journaling to assess STEM identity, belonging, and skill development. External evaluators reviewed the faculty focus groups to assess mentoring effectiveness. Initial findings show strong faculty engagement with the ARG model, with many adopting adaptive mentoring strategies that enhance student support. Students report increased confidence and belonging within their disciplines. However, cross-disciplinary collaboration remains limited, highlighting the need for more intentional networking within the cohort. Students also emphasized the value of peer collaboration alongside faculty mentorship. These results suggest that undergraduate research can serve as a powerful tool for building community and supporting persistence in STEM. Ongoing efforts will focus on expanding networking opportunities, strengthening peer collaboration, and evaluating long-term impacts on student retention. 

As conventional electronic materials approach their physical limits, the application of ultrafast optical fields to access transient states of matter cap- tures imagination. The inversion symmetry governs the optical parity selection rule, differentiating between accessible and inaccessible states of matter. To circumvent parity-forbidden transitions, the common practice is to break the inversion symmetry by material design or external fields. Here we report how the application of femtosecond ultraviolet pulses can energize a parity-forbidden dark exciton state in black phosphorus while maintaining its intrinsic material symmetry. Unlike its conventional bandgap absorption in visible-to-infrared, femtosecond ultraviolet excitation turns on efficient Coulomb scattering, promoting carrier multiplication and electronic heating to ~3000 K, and consequently populating its parity-forbidden states. Interfero- metric time- and angle-resolved two-photon photoemission spectroscopy reveals dark exciton dynamics of black phosphorus on ~100 fs time scale and its anisotropic wavefunctions in energy-momentum space, illuminating its potential applications in optoelectronics and photochemistry under ultraviolet optical excitation. 

Abstract Global environmental change is causing a decline in biodiversity with profound implications for ecosystem functioning and stability. It remains unclear how global change factors interact to influence the effects of biodiversity on ecosystem functioning and stability. Here, using data from a 24-year experiment, we investigate the impacts of nitrogen (N) addition, enriched CO₂(eCO₂), and their interactions on the biodiversity-ecosystem functioning relationship (complementarity effects and selection effects), the biodiversity-ecosystem stability relationship (species asynchrony and species stability), and their connections. We show that biodiversity remains positively related to both ecosystem productivity (functioning) and its stability under N addition and eCO₂. However, the combination of N addition and eCO₂diminishes the effects of biodiversity on complementarity and selection effects. In contrast, N addition and eCO₂do not alter the relationship between biodiversity and either species asynchrony or species stability. Under ambient conditions, both complementarity and selection effects are negatively related to species asynchrony, but neither are related to species stability; these links persist under N addition and eCO₂. Our study offers insights into the underlying processes that sustain functioning and stability of biodiverse ecosystems in the face of global change.