Research map / Learning theory and algorithms / Sample complexity
Just Wing It: Near-Optimal Estimation of Missing Mass in a Markovian Sequence
Ashwin Pananjady, Vidya Muthukumar, Andrew Thangaraj
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| use window size tau greater than 1 instead of vanilla Good-Turing (tau = 1) (ablation) | WingIt · Sticky Markov chain (p = 0.5) | MSE decay with n: tau = 4 constant bias; tau = 8 MSE decays in n | Improved |
| use window size tau greater than 1 on i.i.d. data (ablation) | WingIt · IID Uniform([1.2n]) | MSE | No clear effect |
| use data-tuned window size instead of best fixed window (ablation) | WingIt · Sticky Markov chain (p-bar 0.1) | MSE | No clear effect |
| use data-tuned window size instead of best fixed window on stickier chains (ablation) | WingIt · Sticky Markov chain (p-bar 0.9) | MSE | Hurt |
| vary window size tau on non-Markov natural text (ablation) | WingIt · Tale of Two Cities text | MSE | No clear effect |
Closest papers by meaning
- Batch Stationary Distribution Estimation (ICML 2020 · similarity 0.87)
- Estimation of Markov Chain via Rank-Constrained Likelihood (ICML 2018 · similarity 0.87)
- Infinity Learning: Learning Markov Chains from Aggregate Steady-State Observations (AAAI 2020 · similarity 0.86)
- The Collusion of Memory and Nonlinearity in Stochastic Approximation With Constant Stepsize (NeurIPS 2024 · similarity 0.84)
- Foundation Inference Models for Markov Jump Processes (NeurIPS 2024 · similarity 0.84)
- Multi-Marginal Couplings for Metropolis--Hastings (NeurIPS 2026 · similarity 0.84)
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.
