TrAC Self-Supervised Learning
The Self-Supervised Learning in Computer Vision (SSL) digital badge further develops AI skills for creating more efficient and scalable computer vision models. The course is centered on understanding and applying popular SSL methods like SimCLR, MoCo, BYOL, and Vision Transformers (DINO) utilizing PyTorch. The SSL course is offered by Iowa State University's Translational AI Center (TrAC) and is part of a larger Advanced AI Techniques pathway program.
Skills / Knowledge
- Artificial Intelligence
- Problem-solving
- Self-Supervised Learning
Earning Criteria
Required
The Self-Supervised Learning in Computer Vision (SSL) Badge is earned after successful completion of a 4-week, asynchronous, self-paced online course consisting of 5 modules. The 5 modules cover popular SSL methods such as SimCLR, MoCo, BYOL, and Vision Transformers (DINO), and get hands-on experience using PyTorch to build these models.
This course offers a blend of hands-on activities, assignments, video lectures and tutorials.
Learning Outcomes:
Explain the principles of self-supervised learning (SSL) and its importance in computer vision
Describe classical SSL methods and their applications in computer vision
Analyze recent advancements in SSL, focusing on state-of-the-art methods and their underlying principles
Implement contrastive learning methods and understand their role in self-supervised representation learning
Apply clustering-based and Vision Transformer SSL methods to different computer vision tasks
Evaluate the performance and effectiveness of various SSL models on benchmark datasets
Fine-tune self-supervised learning models for downstream tasks like image classification, object detection, and segmentation
Explore advanced SSL techniques and their applications to more complex tasks beyond standard vision problems
Understand future trends in SSL research, such as multi-modal SSL and unsupervised reinforcement learning
Evaluate the potential impact of emerging SSL methods on different domains and industries
Assessment:
Participants will be assessed on:
Engagement with each module
Multiple SSL Coding exercises
SSL Project addressing downstream tasks and evaluating model performance
1 Quiz on SSL concepts, classical methods, and recent advancements in computer vision
About TrAC
The Translational AI Center will break down disciplinary silos to bring together core Iowa State artificial intelligence researchers and subject matter experts interested in applying new technologies to their work. The center will initially focus on conducting core artificial intelligence research, as well as pursuing five application areas of artificial intelligence:
Materials design and manufacturing
Biology, healthcare, and quality of life
Autonomy, intelligent transportation, and smart infrastructure
Food, energy, and water
Ethics, fairness, and adoption.
In addition to serving as a scientific hub for translational artificial intelligence, the center will organize research seminars, host workshops, training, and onboarding programs, offer seed funding for research projects, and serve as an intermediary between private industry partners seeking research services and appropriate university faculty.