Alimama Tech
Aug 23, 2023 · Artificial Intelligence
Reinforcement Learning for Pacing in Preloaded Ads (RLTP)
The paper introduces RLTP, a reinforcement‑learning‑based pacing system that models delayed‑impression preloaded ads as an MDP, uses a dueling DQN to select traffic probabilities, and simultaneously meets exposure targets, ensures smooth delivery, and maximizes CTR, outperforming rule‑based and PID baselines while removing complex multi‑stage pipelines.
RLTPad pacingdelayed impression
0 likes · 16 min read