model misspecification
Occurring when the chosen model does not accurately represent the underlying data generating process. This can lead to poor generalization performance and unreliable predictions.
- Distributional Adversarial Attacks and Training in Deep Hedging
- Doubly Robust Alignment for Large Language Models
- Effortless, Simulation-Efficient Bayesian Inference using Tabular Foundation Models
- How Patterns Dictate Learnability in Sequential Data
- Inductive Domain Transfer In Misspecified Simulation-Based Inference
- Inverse Methods for Missing Data Imputation
- Robust Reinforcement Learning in Finance: Modeling Market Impact with Elliptic Uncertainty Sets
- Smooth Sailing: Lipschitz-Driven Uncertainty Quantification for Spatial Associations
- Tractable Multinomial Logit Contextual Bandits with Non-Linear Utilities