arXiv subjects

Aarthy Nagarajan

Publications and source records attributed to Aarthy Nagarajan.

5 recordsLinked to original sources

arXiv2026

NEXUS-MI: Communication-Aware Federated Personalization for Gateway-Coordinated Motor-Imagery Brain-Computer Interfaces

Electroencephalography (EEG)-based motor-imagery brain-computer interfaces (MI-BCIs) vary across subjects and sessions, complicating personalization from limited calibration data. Federated learning can exploit shared representations without centralizing raw EEG, but existing federated MI studies largely assume regular synchronization. We introduce NEXUS-MI, a gateway-coordinated federated personalization framework that treats synchronization as a coupled learning-and-communication control problem. Raw EEG and classifier heads remain local, while an edge coordinator maintains the shared backbone. We evaluate NEXUS-MI through offline replay using BCI Competition IV Dataset 2a (BCICIV-2a; 9 subjects, 4 classes) and OpenBMI (54 subjects, 2 classes). Session 1 supports backbone learning, and Session 2 provides limited-calibration personalization and held-out testing. An ideal-link reference and six heterogeneous-link policies characterize gateway participation, buffering, stale-update admission, and backbone-download control. The principal comparison holds delayed-update handling fixed while contrasting non-adaptive and communication-aware synchronization. Paired subject-level comparisons use Holm adjustment, and robustness across five matched realizations is assessed by hierarchical bootstrap. Communication-aware coordination reduced server-to-client backbone traffic by approximately 42% on both datasets, while cohort-level accuracy differences were small and realization-dependent. Cohort averages also concealed subject-level vulnerability, with losses reaching approximately 12 percentage points on BCICIV-2a relative to the ideal-link reference. These findings establish gateway synchronization as an explicit design variable in federated MI personalization and motivate joint evaluation of personalized accuracy, communication cost, update freshness, and subject-level reliability.

cs.LG↗

arXiv2026

Slow-Fast Brain-Computer Interfaces: Preventing Neuroadaptive Overfitting in AI-Mediated Neural Interfaces

Artificial intelligence (AI) is transforming brain-computer interfaces (BCIs) from task-specific neural decoders into adaptive systems that complete language, smooth movement, regulate rehabilitation support and adjust stimulation. These capabilities can increase speed, fluency, usability and clinical reach, yet conventional performance metrics may overlook losses in intent fidelity, authorship, agency, therapeutic challenge and durable clinical benefit. I define neuroadaptive overfitting as a closed-loop failure mode in which an AI-mediated BCI becomes over-optimized to short-term proxies of success, including reduced effort, rapid acceptance, lower workload or smooth task completion, while drifting from the user's durable goals. I then propose Slow-Fast BCI, a framework for pacing AI assistance according to decoder evidence, uncertainty, contextual and clinical stakes, fatigue, and user- or clinician-defined goals. The framework distinguishes fast assistance when intent is clear and stakes are low, guarded assistance under uncertainty and slow assistance when misalignment could compromise safety, agency, authorship, motor learning or therapeutic value. Across communication, motor-control, neurorehabilitation and closed-loop neuromodulation applications, I outline corresponding safeguards and evaluation measures. This Perspective argues that AI-mediated BCIs should be evaluated not only by decoding accuracy and task performance, but also by how AI assistance is deployed: when systems act autonomously, seek confirmation, preserve user effort or return control to the user.

q-bio.NC↗

arXiv2026

Compute Allocation in Evolutionary Search: From Depth-Breadth to Multi-Armed Bandits

LLM-guided evolutionary search (Evolve systems) has reached state-of-the-art results on mathematical and combinatorial tasks, yet most existing systems report only the best of many runs and leave the run-to-run distribution undocumented. We ask how a fixed budget of LLM calls should be allocated, and how reliably a single run reaches the reported numbers. Sweeping the depth-breadth grid over five models and three tasks, we identify two empirical regularities: a fitness-compute envelope along which capability ordering largely collapses on effective FLOPs, and a bilinear depth-breadth fit with task-specific interaction; both are gated by model-task capability. Motivated by these regularities, we propose BaSE (Bandit-based Self-Evolving), a multi-armed bandit that allocates LLM calls across parallel trajectories. Without changing the model, prompt, or evaluator, BaSE improves mean fitness by 12.3% over the strongest island-protocol baseline across 8 (model, task) cells, with the largest gains on high-variance settings: a reliability gain from allocation alone.

cs.CL↗

arXiv2026

Functional Connectivity-Guided Band Selection for Motor Imagery Brain-Computer Interfaces

Reliable control in motor imagery brain-computer interfaces (MI-BCIs) requires the precise decoding of user-specific neural rhythms, which vary significantly across individuals. The Common Spatial Pattern (CSP) algorithm is a cornerstone of MI-BCI decoding, yet its performance depends strongly on the spectral range of the input EEG data. Although Filter Bank CSP (FBCSP) extends this as a data-driven decoding framework, its frequency sub-bands are predefined rather than selected using subject-specific physiological criteria. This paper presents a proof-of-concept study of static functional connectivity (FC)-guided band selection for MI-BCI, demonstrated using a conventional FBCSP-based pipeline. The proposed method identifies the most discriminative spectral bands by calculating phase-based connectivity across four sensorimotor channels using wPLI, PLV, and PLI. Nine bands in a 4-40 Hz filter bank are ranked by the effect size of their hemispheric coupling differences and pruned to the top K bands for feature extraction and classification via FBCSP and a Support Vector Regressor. This framework was tested for K values ranging from 1 to 8 across the BCI Competition IV-2a (n = 9) and OpenBMI (n = 54) datasets. Performance was benchmarked against standard nine-band FBCSP and random ablation to determine the minimum number of bands (K*) required to maintain accuracy within a 2% baseline equivalence zone. Results show FC-guided selection can outperform random ablation and achieve near-baseline performance while reducing required CSP fits by 22.2% to 77.8%. PLV enables the most aggressive dimensionality reduction by prioritizing the μ and low-\b{eta} ranges, while wPLI demonstrates superior inter-session robustness by mitigating volume conduction. These findings establish FC-guided selection as a principled and interpretable alternative to heuristic filter bank designs.

q-bio.NC↗

arXiv2019

Understanding the Twitter Usage of Science Citation Index (SCI) Journals

This paper investigates the Twitter interaction patterns of journals from the Science Citation Index (SCI) of Master Journal List (MJL). A total of 953,253 tweets extracted from 857 journal accounts, were analyzed in this study. Findings indicate that SCI journals interacted more with each other but much less with journals from other citation indices. The network structure of the communication graph resembled a tight crowd network, with Nature journals playing a major part. Information sources such as news portals and scientific organizations were mentioned more in tweets, than academic journal Twitter accounts. Journals with high journal impact factors (JIFs) were found to be prominent hubs in the communication graph. Differences were found between the Twitter usage of SCI journals with Humanities and Social Sciences (HSS) journals.

cs.DL↗