arXiv subjects

Fariha Tasmin

Publications and source records attributed to Fariha Tasmin.

5 recordsLinked to original sources

arXiv2025

Constraint Satisfaction Approaches to Wordle: Novel Heuristics and Cross-Lexicon Validation

Wordle presents an algorithmically rich testbed for constraint satisfaction problem (CSP) solving. While existing solvers rely on information-theoretic entropy maximization or frequency-based heuristics without formal constraint treatment, we present the first comprehensive CSP formulation of Wordle with novel constraint-aware solving strategies. We introduce CSP-Aware Entropy, computing information gain after constraint propagation rather than on raw candidate sets, and a Probabilistic CSP framework integrating Bayesian word-frequency priors with logical constraints. Through evaluation on 2,315 English words, CSP-Aware Entropy achieves 3.54 average guesses with 99.9% success rate, a statistically significant 1.7% improvement over Forward Checking (t=-4.82, p<0.001, Cohen's d=0.07) with 46% faster runtime (12.9ms versus 23.7ms per guess). Under 10% noise, CSP-aware approaches maintain 5.3 percentage point advantages (29.0% versus 23.7%, p=0.041), while Probabilistic CSP achieves 100% success across all noise levels (0-20%) through constraint recovery mechanisms. Cross-lexicon validation on 500 Spanish words demonstrates 88% success with zero language-specific tuning, validating that core CSP principles transfer across languages despite an 11.2 percentage point gap from linguistic differences (p<0.001, Fisher's exact test). Our open-source implementation with 34 unit tests achieving 91% code coverage provides reproducible infrastructure for CSP research. The combination of formal CSP treatment, constraint-aware heuristics, probabilistic-logical integration, robustness analysis, and cross-lexicon validation establishes new performance benchmarks demonstrating that principled constraint satisfaction techniques outperform classical information-theoretic and learning-based approaches for structured puzzle-solving domains.

cs.CL↗

arXiv2025

Beyond Static Knowledge Messengers: Towards Adaptive, Fair, and Scalable Federated Learning for Medical AI

Medical AI faces challenges in privacy-preserving collaborative learning while ensuring fairness across heterogeneous healthcare institutions. Current federated learning approaches suffer from static architectures, slow convergence (45-73 rounds), fairness gaps marginalizing smaller institutions, and scalability constraints (15-client limit). We propose Adaptive Fair Federated Learning (AFFL) through three innovations: (1) Adaptive Knowledge Messengers dynamically scaling capacity based on heterogeneity and task complexity, (2) Fairness-Aware Distillation using influence-weighted aggregation, and (3) Curriculum-Guided Acceleration reducing rounds by 60-70%. Our theoretical analysis provides convergence guarantees with epsilon-fairness bounds, achieving O(T^{-1/2}) + O(H_max/T^{3/4}) rates. Projected results show 55-75% communication reduction, 56-68% fairness improvement, 34-46% energy savings, and 100+ institution support. The framework enables multi-modal integration across imaging, genomics, EHR, and sensor data while maintaining HIPAA/GDPR compliance. We propose MedFedBench benchmark suite for standardized evaluation across six healthcare dimensions: convergence efficiency, institutional fairness, privacy preservation, multi-modal integration, scalability, and clinical deployment readiness. Economic projections indicate 400-800% ROI for rural hospitals and 15-25% performance gains for academic centers. This work presents a seven-question research agenda, 24-month implementation roadmap, and pathways toward democratizing healthcare AI.

cs.CY↗

arXiv2025

Detecting and Preventing Latent Risk Accumulation in High-Performance Software Systems

Modern distributed systems employ aggressive optimization strategies that create latent risks - hidden vulnerabilities where exceptional performance masks catastrophic fragility when optimizations fail. Cache layers achieving 99% hit rates can obscure database bottlenecks until cache failures trigger 100x load amplification and cascading collapse. Current reliability engineering focuses on reactive incident response rather than proactive detection of optimization-induced vulnerabilities. This paper presents the first comprehensive framework for systematic latent risk detection, prevention, and optimization through integrated mathematical modeling, intelligent perturbation testing, and risk-aware performance optimization. We introduce the Latent Risk Index (LRI) that correlates strongly with incident severity (r=0.863, p<0.001), enabling predictive risk assessment. Our framework integrates three systems: HYDRA employing six optimization-aware perturbation strategies achieving 89.7% risk discovery rates, RAVEN providing continuous production monitoring with 92.9% precision and 93.8% recall across 1,748 scenarios, and APEX enabling risk-aware optimization maintaining 96.6% baseline performance while reducing latent risks by 59.2%. Evaluation across three testbed environments demonstrates strong statistical validation with large effect sizes (Cohen d>2.0) and exceptional reproducibility (r>0.92). Production deployment over 24 weeks shows 69.1% mean time to recovery reduction, 78.6% incident severity reduction, and 81 prevented incidents generating 1.44M USD average annual benefits with 3.2-month ROI. Our approach transforms reliability engineering from reactive incident management to proactive risk-aware optimization.

cs.SE↗

arXiv2025

Next-Generation Event-Driven Architectures: Performance, Scalability, and Intelligent Orchestration Across Messaging Frameworks

Modern distributed systems demand low-latency, fault-tolerant event processing that exceeds traditional messaging architecture limits. While frameworks including Apache Kafka, RabbitMQ, Apache Pulsar, NATS JetStream, and serverless event buses have matured significantly, no unified comparative study evaluates them holistically under standardized conditions. This paper presents the first comprehensive benchmarking framework evaluating 12 messaging systems across three representative workloads: e-commerce transactions, IoT telemetry ingestion, and AI inference pipelines. We introduce AIEO (AI-Enhanced Event Orchestration), employing machine learning-driven predictive scaling, reinforcement learning for dynamic resource allocation, and multi-objective optimization. Our evaluation reveals fundamental trade-offs: Apache Kafka achieves peak throughput (1.2M messages/sec, 18ms p95 latency) but requires substantial operational expertise; Apache Pulsar provides balanced performance (950K messages/sec, 22ms p95) with superior multi-tenancy; serverless solutions offer elastic scaling for variable workloads despite higher baseline latency (80-120ms p95). AIEO demonstrates 34\% average latency reduction, 28\% resource utilization improvement, and 42% cost optimization across all platforms. We contribute standardized benchmarking methodologies, open-source intelligent orchestration, and evidence-based decision guidelines. The evaluation encompasses 2,400+ experimental configurations with rigorous statistical analysis, providing comprehensive performance characterization and establishing foundations for next-generation distributed system design.

cs.DC↗

arXiv2025

Feature Selection and Regularization in Multi-Class Classification: An Empirical Study of One-vs-Rest Logistic Regression with Gradient Descent Optimization and L1 Sparsity Constraints

Multi-class wine classification presents fundamental trade-offs between model accuracy, feature dimensionality, and interpretability - critical factors for production deployment in analytical chemistry. This paper presents a comprehensive empirical study of One-vs-Rest logistic regression on the UCI Wine dataset (178 samples, 3 cultivars, 13 chemical features), comparing from-scratch gradient descent implementation against scikit-learn's optimized solvers and quantifying L1 regularization effects on feature sparsity. Manual gradient descent achieves 92.59 percent mean test accuracy with smooth convergence, validating theoretical foundations, though scikit-learn provides 24x training speedup and 98.15 percent accuracy. Class-specific analysis reveals distinct chemical signatures with heterogeneous patterns where color intensity varies dramatically (0.31 to 16.50) across cultivars. L1 regularization produces 54-69 percent feature reduction with only 4.63 percent accuracy decrease, demonstrating favorable interpretability-performance trade-offs. We propose an optimal 5-feature subset achieving 62 percent complexity reduction with estimated 92-94 percent accuracy, enabling cost-effective deployment with 80 dollars savings per sample and 56 percent time reduction. Statistical validation confirms robust generalization with sub-2ms prediction latency suitable for real-time quality control. Our findings provide actionable guidelines for practitioners balancing comprehensive chemical analysis against targeted feature measurement in resource-constrained environments.

cs.LG↗