Research Vision

Advancing Distributed, Collaborative, and Trustworthy Intelligence across the Computing Continuum

My research investigates the foundations, architectures, and technologies required to enable intelligent systems to operate collaboratively, efficiently, and securely across distributed computing environments.

As artificial intelligence becomes increasingly integrated into everyday devices, critical infrastructure, industrial systems, and cloud services, traditional centralized approaches face growing challenges in scalability, privacy, communication efficiency, energy consumption, and system autonomy.

My research addresses these challenges by exploring how intelligence can be distributed, coordinated, optimized, and managed across the Edge–Cloud Continuum.

I combine approaches from artificial intelligence, machine learning, distributed systems, data engineering, and systems optimization to design adaptive computing architectures capable of supporting the next generation of intelligent applications.

My research portfolio is organized into six interconnected research lines.

Research Line 1: Federated and Collaborative Machine Learning

Keywords: Federated LearningDistributed TrainingCollaborative AINon-IID DataModel AggregationFederated Optimization

Federated and collaborative machine learning is a central component of my research portfolio.

I investigate how multiple distributed participants can collaboratively develop intelligent models without requiring centralized access to their private or locally generated data.

My research addresses fundamental challenges associated with data heterogeneity, resource-constrained participants, communication overhead, unreliable connectivity, model convergence, and distributed optimization.

Particular emphasis is placed on designing learning strategies that account for the capabilities, availability, and contributions of participating devices.

Research directions:

  • Federated optimization and adaptive aggregation strategies
  • Learning under statistical and system heterogeneity
  • Communication-efficient and resource-aware federated learning
  • Participant selection, scheduling, and utility-aware collaboration
  • Federated knowledge distillation and model personalization
  • Robustness against participant churn and concept drift
  • Federated learning for foundation models and large language models
  • Collaborative learning across cross-device and cross-silo environments

Research objective: To develop scalable, adaptive, and efficient collaborative learning frameworks that support heterogeneous participants and diverse real-world environments.

Research Line 2: Edge Intelligence and Distributed AI Systems

Keywords: Edge AIEdge–Cloud ContinuumDistributed IntelligenceAI InferenceResource ManagementNeuromorphic Computing

This research line investigates how intelligent computing capabilities can be deployed and coordinated across edge devices, local infrastructure, and cloud platforms.

The increasing deployment of AI in latency-sensitive and resource-constrained environments requires architectures that balance computational performance, energy efficiency, privacy, and communication costs.

My work explores distributed intelligence as a system-level capability, where learning, inference, data processing, and resource allocation are coordinated dynamically across heterogeneous infrastructure.

Research directions:

  • Edge-native and distributed AI architectures
  • Adaptive placement of AI workloads across edge and cloud
  • Lightweight machine learning and model compression
  • Resource-aware distributed inference and training
  • Communication-efficient intelligent systems
  • Spiking neural networks and neuromorphic edge intelligence
  • AI-enabled resource management and workload scheduling
  • Performance optimization in heterogeneous computing environments

Research objective: To enable intelligent applications to operate reliably and efficiently across heterogeneous edge–cloud environments.

Research Line 3: Trustworthy, Privacy-Preserving, and Secure AI

Keywords: Trustworthy AIPrivacy-Preserving LearningZero TrustSecure AggregationFederated SecurityAI Robustness

Trust, privacy, and security are essential requirements for distributed intelligent systems, particularly when multiple independent organizations, devices, and stakeholders collaborate.

My research explores methods for protecting sensitive information, strengthening trust between distributed participants, and improving the security and robustness of collaborative learning systems.

This includes investigating security threats arising from malicious participants, information leakage, unreliable model updates, and heterogeneous trust assumptions.

Research directions:

  • Privacy-preserving distributed and federated learning
  • Zero-trust architectures for collaborative AI
  • Secure aggregation and distributed trust management
  • Differential privacy and privacy–utility trade-offs
  • Robustness against poisoning and adversarial attacks
  • Secure and trustworthy model sharing
  • Federated unlearning and responsible data governance
  • Trustworthy AI in healthcare and critical infrastructure

Research objective: To develop collaborative AI systems that provide strong privacy and security guarantees while maintaining practical learning performance and operational efficiency.

Research Line 4: Agentic AI and Autonomous Distributed Systems

Keywords: Agentic AIMulti-Agent SystemsAutonomous OrchestrationDistributed Decision-MakingIntelligent Agents

Emerging AI systems are evolving from passive prediction models into autonomous agents capable of reasoning, planning, coordinating, and acting within complex environments.

This transformation creates opportunities for intelligent computing infrastructures that can manage themselves, adapt to operational conditions, and coordinate decisions across distributed resources.

My research explores the integration of agentic AI with distributed systems, particularly the design of autonomous orchestration and decentralized decision-making mechanisms.

Research directions:

  • Agentic AI architectures for distributed computing
  • Multi-agent coordination across the edge–cloud continuum
  • Autonomous workload placement and resource orchestration
  • Decentralized planning and delegated decision-making
  • Small language models and distributed AI agents
  • AI-driven monitoring, adaptation, and fault recovery
  • Proactive and reactive orchestration under uncertainty
  • Agent-based management of heterogeneous computing infrastructure

Research objective: To advance autonomous distributed systems that can intelligently plan, coordinate, and adapt their operations with limited human intervention.

Research Line 5: Sustainable, Resilient, and Resource-Efficient Computing

Keywords: Green AISustainable ComputingEnergy-Aware OrchestrationResilienceResource OptimizationClimate-Aware Systems

The growth of AI workloads and distributed computing infrastructure introduces significant challenges in energy consumption, resource utilization, operational resilience, and environmental sustainability.

My research investigates how intelligent resource management and adaptive orchestration can improve the efficiency and resilience of computing systems.

An important direction is the development of proactive and reactive mechanisms that respond to changes in workload demand, energy availability, network conditions, and infrastructure disruptions.

Research directions:

  • Energy-aware AI workload scheduling and placement
  • Sustainable federated learning and distributed inference
  • Climate-aware computing and infrastructure orchestration
  • Resilient distributed systems under power and network disruptions
  • Predictive modeling of workload and resource demand
  • Adaptive resource allocation and capacity planning
  • Performance–energy–reliability trade-offs
  • Simulation-based evaluation of resilient computing systems

Research objective: To design intelligent computing infrastructures that minimize unnecessary resource consumption while maintaining performance, adaptability, and resilience.

Research Line 6: Data-Intensive Systems and Cloud-Native Computing

Keywords: Big DataDistributed Data SystemsCloud ComputingData EngineeringStreaming SystemsCloud-Native Infrastructure

My background in big data and cloud computing provides an important systems foundation for my research in distributed intelligence.

This research line focuses on the design, management, and performance evaluation of data-intensive computing systems that support large-scale analytics, machine learning, and intelligent applications.

I investigate how data processing, storage, communication, and computing infrastructure can be optimized to meet demanding performance, scalability, and reliability requirements.

Research directions:

  • Big data resource management and optimization
  • Distributed data processing and analytics
  • Cloud-native architectures and container orchestration
  • Streaming data systems and real-time analytics
  • Distributed storage and data management
  • Performance benchmarking and systems observability
  • Elastic scaling and workload prediction
  • Data infrastructure for AI and machine learning applications

Research objective: To build scalable, reliable, and efficient computing foundations for data-intensive and AI-driven applications.

Cross-Cutting Research Themes

The six research lines are connected through several recurring scientific challenges:

Heterogeneity and Adaptation: Enabling systems to operate across participants and infrastructures with different data distributions, capabilities, and operating conditions.

Efficiency and Optimization: Improving computational efficiency, communication performance, resource utilization, and energy consumption.

Privacy, Security, and Trust: Embedding privacy protection, robustness, and trust into the foundations of distributed intelligence.

Autonomy and Collaboration: Supporting intelligent decision-making and coordination across decentralized environments.

Scalability and Resilience: Ensuring that intelligent systems remain dependable as their scale, complexity, and operational conditions change.

Research Applications and Impact

My research is motivated by both fundamental scientific questions and the practical requirements of emerging intelligent applications.

Representative application areas include:

  • Smart Industry and Industry 4.0: Collaborative intelligence for industrial automation, predictive analytics, and distributed decision-making.
  • Digital Healthcare: Privacy-preserving machine learning and secure collaboration across healthcare institutions.
  • Smart Energy and Sustainability: Distributed forecasting, intelligent resource management, and energy-aware computing.
  • Autonomous and Cyber-Physical Systems: Adaptive intelligence for interconnected and resource-constrained environments.
  • Cloud and Edge Infrastructure: Intelligent orchestration, distributed resource management, and resilient computing platforms.
  • Next-Generation AI Services: Collaborative foundation models, agentic systems, and distributed generative AI.

Research Collaboration

I actively welcome interdisciplinary research collaboration with academic institutions, industry partners, and international research organizations.

I am particularly interested in partnerships addressing federated and distributed AI, edge intelligence, autonomous computing, trustworthy machine learning, and sustainable intelligent infrastructure.

My research activities encompass fundamental research, applied systems development, experimental evaluation, postgraduate supervision, and collaborative research initiatives.

The overarching goal of my research is to make intelligent computing systems more collaborative, autonomous, trustworthy, efficient, and sustainable.