Machine Learning Algorithm Laboratory
Embedded Files
Lab Shorts최종.mp4Welcome to the Machine Learning Algorithm Laboratory!
Our lab focuses on designing algorithms that enhance the generalization ability of deep neural networks and applying them to complex real-world problems. We are particularly interested in exploring and leveraging uncertainty in deep learning, as we believe it holds the key to building robust and adaptive AI systems.
At MLALab, we aim to bridge algorithmic theory and practical deployment, pushing the frontier of intelligent systems that learn efficiently and perform reliably in uncertain environments. If you are interested in artificial intelligence and its application, please send me an email with your short transcript and CV, to yykim@uos.ac.kr.
🎯 Core Research Themes
🔍 Generalization in Deep Learning
We investigate foundational aspects that enable models to generalize beyond their training distributions:
Active Learning: Selecting informative samples to reduce labeling cost.
Out-of-Distribution (OOD) Discovery: Detecting and characterizing unseen or anomalous data.
Data Augmentation: Synthesizing diverse training data to enhance model robustness.
Domain Generalization: Transferring knowledge across domains with distribution shifts.
🛠️ Real-world Applications
We apply our algorithms to high-impact tasks that require reliability and precision:
Filtering and Target Tracking: Following dynamic objects in noisy environments.
Object Detection: Identifying and localizing instances in visual scenes.
Semantic Segmentation: Understanding the structure of images at pixel-level granularity.
Page updated
Google Sites
Report abuse