prior knowledge
Existing understandings, theories, or information that inform the development and training of AI models, helping to improve learning efficiency, reduce the data required for training, and enhance model generalization.
- Anytime-valid, Bayes-assisted, Prediction-Powered Inference
- Energy Loss Functions for Physical Systems
- KINDLE: Knowledge-Guided Distillation for Prior-Free Gene Regulatory Network Inference
- Less Greedy Equivalence Search
- Non-Stationary Lipschitz Bandits
- Pattern-Guided Adaptive Prior for Structure Learning
- PhysGym: Benchmarking LLMs in Interactive Physics Discovery with Controlled Priors
- Raw2Drive: Reinforcement Learning with Aligned World Models for End-to-End Autonomous Driving (in CARLA v2)
- Segment Anything Model Meets Semi-supervised Medical Image Segmentation: A Novel Perspective