distributional shift
Distributional shift refers to changes in the data distribution of input features that an AI model encounters between the training phase and deployment. Such shifts can degrade model performance, making it crucial to develop methods that can handle or adapt to these changes.
- Delta Attention: Fast and Accurate Sparse Attention Inference by Delta Correction
- Improving Generalization of Neural Combinatorial Optimization for Vehicle Routing Problems via Test-Time Projection Learning
- Pessimistic Data Integration for Policy Evaluation
- Rebalancing Return Coverage for Conditional Sequence Modeling in Offline Reinforcement Learning
- SVRPBench: A Realistic Benchmark for Stochastic Vehicle Routing Problem