Upload README.md

README.md

@@ -2,6 +2,10 @@
 
 This repository contains the results of comparing different bidding algorithms for Real-Time Bidding (RTB) in online advertising, optimizing for clicks under budget constraints.
 
+**📚 NEW: Complete research resources are now available:**
+- **[RESEARCH_RESOURCES.md](RESEARCH_RESOURCES.md)** — Full literature survey covering 32 papers across bidding algorithms, CTR prediction, and clearing price models
+- **[AUDIT_TRAIL.md](AUDIT_TRAIL.md)** — Every paper, dataset, codebase, and external resource consulted (44 total)
+
 ## Problem Setup
 
 - **Objective**: Maximize number of clicks
@@ -53,10 +57,32 @@ This repository contains the results of comparing different bidding algorithms f
 - `train.py` — Full training and comparison script
 - `results.json` — Synthetic data results
 - `results_real.json` — Real Criteo data results
+- `RESEARCH_RESOURCES.md` — Complete literature survey (32 papers)
+- `AUDIT_TRAIL.md` — Every resource consulted (44 items)
+
+## Recommended Next Steps (from research)
+
+1. **Upgrade CTR model**: Replace LogisticRegression with FinalMLP (AAAI 2023, Criteo AUC 0.8149)
+2. **Add clearing price model**: Use TorchSurv for censored regression
+3. **Add Balseiro dual mirror descent**: Second-price baseline for comparison
+4. **Two-sided budget constraint**: Add spend floor (k% minimum) with second dual variable
+5. **Hyperparameter sweep**: Step size ε, budget fraction k%, value per click
 
 ## References
 
+### Bidding Algorithms
 - Wang et al. (2023): "Learning to Bid in Repeated First-Price Auctions with Budgets" [arXiv:2304.13477](https://arxiv.org/abs/2304.13477)
-- Zhang et al. (2014): "Optimal Real-Time Bidding for Display Advertising" (KDD)
+- Balseiro et al. (2020): "Dual Mirror Descent for Online Allocation" [arXiv:2011.10124](https://arxiv.org/abs/2011.10124)
+- Feng et al. (2022): "Online Bidding for RoS Constrained Advertisers" [arXiv:2208.13713](https://arxiv.org/abs/2208.13713)
 - Cai et al. (2017): "Real-Time Bidding by Reinforcement Learning" [arXiv:1701.02490](https://arxiv.org/abs/1701.02490)
-- Balseiro et al. (2023): "Robust Budget Pacing with a Single Sample" [arXiv:2302.02006](https://arxiv.org/abs/2302.02006)
+- Zhang et al. (2014): "Optimal Real-Time Bidding for Display Advertising" (KDD)
+
+### CTR Prediction
+- Mao et al. (2023): "FinalMLP: Two-Stream MLP for CTR" [arXiv:2304.00902](https://arxiv.org/abs/2304.00902)
+- Wang et al. (2023): "GDCN: Gated Deep Cross Network" [arXiv:2311.04635](https://arxiv.org/abs/2311.04635)
+- Wang et al. (2021): "DCN V2" [arXiv:2008.13535](https://arxiv.org/abs/2008.13535)
+- Zhu et al. (2021): "BARS-CTR Benchmark" [arXiv:2009.05794](https://arxiv.org/abs/2009.05794)
+
+### Clearing Price Prediction
+- Wu et al. (2015): "Predicting Winning Price with Censored Data" (KDD)
+- TorchSurv: Deep Survival Analysis [arXiv:2404.10761](https://arxiv.org/abs/2404.10761)