estimation error
The difference between the estimated value and the true value of a parameter or variable in a model, impacting the overall accuracy and performance.
- A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning
- Active Measurement: Efficient Estimation at Scale
- Adaptive Variance Inflation in Thompson Sampling: Efficiency, Safety, Robustness, and Beyond
- Adversarial Robustness of Nonparametric Regression
- Balancing Positive and Negative Classification Error Rates in Positive-Unlabeled Learning
- Beyond the Seen: Bounded Distribution Estimation for Open-Vocabulary Learning
- Breaking the Order Barrier: Off-Policy Evaluation for Confounded POMDPs
- Evaluating multiple models using labeled and unlabeled data
- Instance-Optimality for Private KL Distribution Estimation
- Language Models Can Predict Their Own Behavior
- Learning Personalized Ad Impact via Contextual Reinforcement Learning under Delayed Rewards
- Sparse MeZO: Less Parameters for Better Performance in Zeroth-Order LLM Fine-Tuning
- State Size Independent Statistical Error Bound for Discrete Diffusion Models
- Treatment Effect Estimation for Optimal Decision-Making