인공지능대학원

Sequential Decision, Multi-armed bandit algorithms, Online learning, Causal inference, Policy evaluation, Causal inference, Missing data analysis

• Inverse problems for various imaging modalities:

- natural image restorations (super-resolution, denoising, deblurring, etc.)

- medical image reconstructions (MRI, CT, SIM, Cryo-EM, DOT, EEG, fMRI, etc.)

• Bridging between signal processing and deep learning communities:

- providing a design principle for deep learning architectures

- network analysis using topological data analysis (TDA) 

• Deep generative models:

- developing a high fidelity and diverse image-to-image translation model

- improving generative models based on theoretical understandings

- Reinforcement Learning for Large Language Models

- Large Language Models & LLM Agen

- Offline Reinforcement Learning & Offline-to-Online Reinforcement Learning

- Safe Reinforcement Learning

- Reinforcement Learning for Industrial AI

Generative modeling, Multi-agent RL, model-based RL, collaboration, cooperation, Bayesian machine learning, natural language processing, HCI, data mining, Graph neural networks, recommender systems, fake news

· 멀티모달 LLM(Multimodal LLM ): Vision-Language Model

· Embodied AI/Vision-Language Action Model

· 텍스트-이미지/비디오 생성(Text-to-image/video generation)

· 멀티모달 대화 모델(Multi-modal conversational models)

· 비디오 이해 및 답변 생성 모델(Video understanding and question answering)

· Computer Vision

· Deep/Machine Learning

· Pose Estimation of Human Body, Face and Hand

· Action and Gesture Recognition

· Semantic Segmentation

· Object Recognition

• Nonstationary Degradation Identification & Restoration
• Design of Efficient Deep Neural Networks & Applications
• Product Quality Inspection & Process Automation

• Dataset Distillation and Coreset Selection
• Re-Identification
• 3D Point Clouds Processing and Analysis
• Person Search
• Image Restoration
• Reflection Removal

그래프 표현 학습, 신뢰할 수 있는 그래프 마이닝, 추천 시스템, 그래프 + X (커리어 모델링, 이상 탐지 등)

Graph Representation Learning, Trustworthy Graph Mining, Recommender Systems, Graph + X (Career Modeling, Anomaly Detection, etc)

• Offline Reinforcement Learning

Offline-RL learns policy by utilizing only the experience it gathers without additional interaction with the environment. There are still some challenges, such as distribution shifts and out-of-distribution problem.

• Domain Adaptation/Imitation Learning

Imitation learning is a branch of research aiming to apply reinforcement learning to real-life scenarios, focusing on learning policies that mimic the actions of experts. Recently, there has been progress in research on Domain Adaptation and Cross-Domain studies, enabling imitation of actions from experts in different domains.

•  Multi-Agent Reinforcement Learning

In multi-agent RL, multiple agents aim to learn policies that would maximize the expected return from a shared environment. Coordination among the agents is essential for achieving this goal as the agents effect themselves as learning progresses.