Researchers have developed HOBA, a novel hierarchical reinforcement learning framework designed to improve online advertising bidding systems. This system decouples strategic reasoning, model selection, and bid execution across different time scales, using a large language model for hyperparameter inference and a SARSA agent for expert model selection. Experiments on the AuctionNet benchmark and a large-scale A/B test showed HOBA outperformed existing methods, leading to a 3.6% increase in target cost in a real-world deployment. AI
IMPACT This framework could lead to more adaptive and efficient online advertising systems by improving bid optimization and reducing manual tuning.
RANK_REASON The cluster contains a research paper detailing a new AI framework for online advertising. [lever_c_demoted from research: ic=1 ai=1.0]
- arXiv
- AuctionNet
- Decision Transformer
- Hugging Face
- large language model
- model predictive control
- offline RL policies
- SARSA agent
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
