Skip to content

index.md

Latest commit

 

History

History
50 lines (36 loc) · 3.15 KB

index.md

File metadata and controls

50 lines (36 loc) · 3.15 KB
layout page
title Agent Intelligent Laboratory
subtitle Agent Intelligent Laboratory, Konkuk University
cover-img /assets/img/server.jpeg

AI Agents · LLM · Multi-Agent Systems · Autonomous Reasoning · Agentic Collaboration · Intelligent Cloud Platforms

About AI Lab

The Agent Intelligent Laboratory (AI Lab) at Konkuk University focuses on cutting-edge research in AI Agent systems — autonomous software entities that perceive, reason, plan, and act to accomplish complex goals. Our mission is to design, build, and deploy intelligent agents that collaborate with humans and other agents to solve real-world problems across diverse domains.

AI Agent & Large Language Model (LLM) Research

We investigate the design and orchestration of LLM-powered autonomous agents capable of multi-step reasoning, tool use, and self-reflection. Our research spans:

  • Agentic AI Architectures — ReAct, Plan-and-Execute, Reflection, and Tree-of-Thought agent frameworks
  • Multi-Agent Collaboration — cooperative and competitive multi-agent systems for complex task decomposition and distributed problem-solving
  • Retrieval-Augmented Generation (RAG) — grounding LLM agents with external knowledge bases for factual and domain-specific responses
  • Tool-Augmented Agents — enabling agents to interact with APIs, databases, code interpreters, and external services autonomously
  • Agent Memory & Planning — long-term memory, hierarchical planning, and experience replay for persistent and adaptive agent behavior

Intelligent Systems & Distributed Platforms

Building on our strong foundation in distributed systems, we develop the infrastructure that powers scalable AI agent deployments:

  • Cloud-Native Agent Platforms — Kubernetes-based orchestration for deploying and scaling multi-agent services
  • Edge-Cloud Cooperative AI — distributed agent inference across fog, edge, and cloud tiers for latency-sensitive applications
  • Digital Twin & Simulation — agent-driven digital twin environments for Urban Air Mobility (UAM) and smart mobility systems
  • Dependability & Performance Engineering — stochastic modeling and evaluation of AI-powered distributed systems

Applied AI & Domain Intelligence

Our agents are applied to impactful real-world domains:

  • Autonomous Navigation — deep reinforcement learning-based agents for mobile robot and UAV path planning
  • Healthcare AI Assistants — multi-modal intelligent agents for elderly care and medical decision support
  • Conversational AI — knowledge-grounded dialog agents and open-domain QA systems powered by LLMs
  • Computer Vision Intelligence — human pose estimation, activity recognition, and visual understanding agents

Core Research Keywords

AI Agent · LLM · Multi-Agent System · Agentic AI · RAG · Tool-Use Agent · Deep Reinforcement Learning · Autonomous Navigation · Digital Twin · Cloud-Native Platform · Conversational AI · Computer Vision

Contact

Office: Konkuk University New Engineering Building 1207-3 (신공학관 1207-3호)
Email: dkmin at konkuk.ac.kr (Prof. Dugki Min, Ph.D.)