Research map / Learning theory and algorithms / Submodular and game algorithms
A Scalable Neural Network for DSIC Affine Maximizer Auction Design
Zhijian Duan, Haoran Sun, Yurong Chen, Xiaotie Deng
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 |
|---|---|---|---|
| Learnable boost lambda instead of fixed lambda=0 (ablation) | AMenuNet · 5x2 (B) | average revenue: 0.3363 -> 0.7548 | Improved |
| Learnable boost lambda instead of fixed lambda=0 (ablation) | AMenuNet · 3x2 (B) | average revenue: 0.3892 -> 0.6197 | Improved |
| Learnable weights w instead of fixed w=1 (ablation) | AMenuNet · 5x2 (B) | average revenue: 0.7010 -> 0.7548 | Improved |
| Learnable weights w instead of fixed w=1 (ablation) | AMenuNet · 2x10 (A) | average revenue: 2.3848 -> 2.3815 | No clear effect |
| Learnable menu instead of all deterministic allocations (A_dtm) (ablation) | AMenuNet · 2x2 (A) | average revenue: 0.3633 -> 0.4293 | Improved |
| Learnable menu instead of all deterministic allocations (A_dtm) (ablation) | AMenuNet · 2x10 (A) and 3x10 (C) | average revenue: 59049 and 1048576 menu entries | Improved |
| Transformer-based architecture instead of 4-layer FCN (ablation) | AMenuNet · 3x10 (C) | average revenue: 5.0262 -> 5.5896 | Improved |
| Transformer-based architecture instead of 4-layer FCN (ablation) | AMenuNet · 2x2 (A) | average revenue: 0.4124 -> 0.4293 | Improved |
| Train AMenuNet on 2x5 and test on 2x10 (out-of-setting) vs Item-Myerson | AMenuNet · 2x10 (A) | average revenue: 2.1398 -> 2.2034 | Improved |
| Train on 2x2 and test on 2x5 (out-of-setting) vs Item-Myerson | AMenuNet · 2x5 (A) | average revenue: 1.0700 -> 1.0503 | Hurt |
| Train on 2x5 and test on 2x2 (smaller than training) | AMenuNet · 2x2 (A) | average revenue: 0.4265 -> 0.1212 | Hurt |
| Affine parameters (AMenuNet) added to VCG | VCG · contextual 3x10 (A) | average revenue: 2.2967 -> 2.9006 | Improved |
Closest papers by meaning
- AuctionNet: A Novel Benchmark for Decision-Making in Large-Scale Games (NeurIPS 2024 · similarity 0.89)
- Learning Optimal Auctions with Correlated Value Distributions (AAAI 2025 · similarity 0.89)
- BundleFlow: Deep Menus for Combinatorial Auctions by Diffusion-Based Optimization (NeurIPS 2025 · similarity 0.87)
- Auctionformer: A Unified Deep Learning Algorithm for Solving Equilibrium Strategies in Auction Games (ICML 2024 · similarity 0.87)
- Simultaneous Optimization of Bid Shading and Internal Auction for Demand-Side Platforms (AAAI 2024 · similarity 0.86)
- Increasing Revenue in Efficient Combinatorial Auctions by Learning to Generate Artificial Competition (AAAI 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.
