Research map / Learning theory and algorithms / Sample complexity
Estimating Maximum Expected Value through Gaussian Approximation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Replace maximum estimator with Weighted Estimator (Gaussian-approximated selection probabilities) | Estimation of max expected value · Internet ads simulation | MSE | Helped |
| Use Weighted Estimator in place of ME and DE | Sponsored search auction incentive compatibility · Sponsored search auction | advertiser utility gain vs bid | Helped |
| Use Weighted Q-Learning (WE for max action value) instead of Q-Learning | Grid world with Bernoulli and Gaussian rewards · Grid world | bias of max Q-value estimate | Helped |
| Use Weighted policy (exploration from WE probabilities) in Weighted Q-Learning (ablation) | Weighted Q-Learning, grid world with N(-1,1) rewards · Grid world | average reward per step | Helped |
| Use Double Q-Learning instead of Q-Learning | Grid world with high reward variance · Grid world | average reward per step | Helped |
| Use Double Q-Learning instead of Q-Learning | Forex MDP · Forex test set | profit per year | Hurt |
| Use Weighted policy in Weighted Q-Learning (ablation) | Weighted Q-Learning, Forex · Forex | profit per year | Mixed |
| Use Bias-corrected Q-Learning instead of Q-Learning | Grid world · Grid world | average reward per step | Mixed |
| Increase number of ads (actions) from 10 to 100 | Maximum estimator vs double estimator · Internet ads | MSE | Mixed |
| Use Weighted Estimator when the click-rate range is wide | WE vs ME · Internet ads, varying click-rate range | MSE | Mixed |
Closest papers by meaning
- Estimating the Maximum Expected Value in Continuous Reinforcement Learning Problems (AAAI 2017 · similarity 0.84)
- Maxmin Q-learning: Controlling the Estimation Bias of Q-learning (ICLR 2020 · similarity 0.84)
- Learning multivariate Gaussians with imperfect advice (ICML 2025 · similarity 0.82)
- Sub-Gaussian Confidence Intervals for Heavy-Tailed Data: Characterizing the Limits of Inference (NeurIPS 2026 · similarity 0.82)
- One Sample Fits All: Approximating All Probabilistic Values Simultaneously and Efficiently (NeurIPS 2024 · similarity 0.81)
- The Bias-Variance Tradeoff in Data-Driven Optimization: A Local Misspecification Perspective (NeurIPS 2025 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.