Research map / Learning theory and algorithms / Combinatorial and search optimization
Contextual Reserve Price Optimization in Auctions via Mixed Integer Programming
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 |
|---|---|---|---|
| Use exact MIP optimization instead of DC | linear contextual reserve-price optimization · Synthetic data, baseline/high-noise/low-correlation/low-margin, train and test | average reward: 0.776 -> 0.962 (baseline train); 0.776 -> 0.924 (baseline test); 0.752 -> 0.907 (high-noise train); 0.750 -> 0.858 (high | Helped |
| Use LP relaxation instead of DC | linear contextual reserve-price optimization · Synthetic data | average reward | Mixed |
| Terminate MIP at the root node rather than continue the MIP search | MIP · Synthetic baseline | average reward: 0.962 pm{0.006} vs 0.962 pm{0.006} (train); 0.924 pm{0.003} vs 0.923 pm{0.002} (test) | No clear effect |
| Use exact MIP optimization instead of DC | linear contextual reserve-price optimization · eBay sports memorabilia, 2000 training samples, train and test | average reward: 0.704 -> 0.726 (train); 0.709 -> 0.714 (test) | Helped |
| Use exact MIP optimization instead of DC | linear contextual reserve-price optimization · eBay sports memorabilia, 5000 training samples, train and test | average reward: 0.704 -> 0.731 (train); 0.704 -> 0.725 (test) | Helped |
| Increase eBay training sample size from 2000 to 5000 | MIP · eBay sports memorabilia, train and test | average reward: 0.726 -> 0.731 (train); 0.714 -> 0.725 (test) | Helped |
| Use LP relaxation instead of exact MIP | linear contextual reserve-price optimization · Worst-case constructed instances as sample size grows | optimal reward bound: 1/n; at least 1 | Hurt |
Closest papers by meaning
- Revenue Optimization with Approximate Bid Predictions (NeurIPS 2017 · similarity 0.89)
- Learning to bid in revenue-maximizing auctions (ICML 2019 · similarity 0.88)
- Learning Optimal Reserve Price against Non-myopic Bidders (NeurIPS 2018 · similarity 0.88)
- Optimal Non-parametric Learning in Repeated Contextual Auctions with Strategic Buyer (ICML 2020 · similarity 0.87)
- Improved Algorithms for Contextual Dynamic Pricing (NeurIPS 2024 · similarity 0.86)
- Procurement Auctions via Approximately Optimal Submodular Optimization (ICML 2025 · similarity 0.86)
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.