Research map / Learning theory and algorithms / Submodular and game algorithms
Benefits of Permutation-Equivariance in Auction Mechanisms
Tian Qin, Fengxiang He, Dingfeng Shi, Wenbing Huang, Dacheng Tao
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 |
|---|---|---|---|
| project RegretNet to permutation-equivariant RegretNet-PE (orbit averaging) | RegretNet · 5x1 Uniform | revenue: 0.658 -> 0.677 | Improved |
| project RegretNet to permutation-equivariant RegretNet-PE | RegretNet · 2x1 Uniform | ex-post regret (x1e5): 17.4 -> 14.6 | Improved |
| project RegretNet to permutation-equivariant RegretNet-PE | RegretNet · 3x1 Uniform | ex-post regret (x1e5): 18.3 -> 16.4 | Improved |
| project RegretNet to permutation-equivariant RegretNet-PE | RegretNet · 5x1 Uniform | generalization error (x1e5): 6.40 -> 5.10 | Improved |
| project RegretNet to permutation-equivariant RegretNet-PE | RegretNet · 3x1 Uniform | generalization error (x1e5): 11.4 -> 10.2 | Improved |
| project RegretNet to permutation-equivariant RegretNet-PE | RegretNet · 2x2 Uniform | revenue: 0.870 -> 0.913 | Improved |
| project well-trained RegretNet post hoc (RegretNet-Test) | RegretNet · 5x1 Uniform | ex-post regret (x1e5): 15.9 -> 6.50 | Improved |
| project well-trained RegretNet post hoc (RegretNet-Test) | RegretNet · 3x1 Uniform | ex-post regret (x1e5): 18.3 -> 13.3 | Improved |
| project well-trained RegretNet post hoc (RegretNet-Test) | RegretNet · 5x1 Uniform | revenue: 0.658 -> 0.658 | No clear effect |
| orbit averaging (RegretNet-PE) instead of parameter-sharing (EquivariantNet) | EquivariantNet · 1x2 Uniform | revenue: 0.551 -> 0.563 | Improved |
| orbit averaging (RegretNet-PE) instead of parameter-sharing (EquivariantNet) | EquivariantNet · 1x2 Uniform | ex-post regret: 0.00013 -> 0.00037 | Hurt |
| orbit averaging (RegretNet-PE) instead of parameter-sharing (EquivariantNet) | EquivariantNet · 2x2 Uniform | ex-post regret: 0.00100 -> 0.00067 | Improved |
Closest papers by meaning
- Learning Optimal Auctions with Correlated Value Distributions (AAAI 2025 · similarity 0.88)
- Beyond Monotonicity: On the Convergence of Learning Algorithms in Standard Auction Games (AAAI 2025 · similarity 0.85)
- Comparing Uniform Price and Discriminatory Multi-Unit Auctions through Regret Minimization (NeurIPS 2025 · similarity 0.85)
- Auctionformer: A Unified Deep Learning Algorithm for Solving Equilibrium Strategies in Auction Games (ICML 2024 · similarity 0.85)
- AuctionNet: A Novel Benchmark for Decision-Making in Large-Scale Games (NeurIPS 2024 · similarity 0.85)
- Safely Learning Optimal Auctions: A Testable Learning Framework for Mechanism Design (ICML 2025 · similarity 0.85)
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.
