Constraint-Aware Generative Auto-bidding via Pareto-Prioritized Regret Optimization

TL;DR

PRO-Bid introduces a novel constraint-aware auto-bidding framework that combines Pareto representation and regret optimization to improve efficiency and constraint satisfaction in marketing campaigns.

Contribution

It proposes a new generative auto-bidding method that effectively decouples constraints and actively optimizes for high utility outcomes, surpassing existing methods.

Findings

Achieves better constraint satisfaction and higher value in experiments.

Outperforms state-of-the-art baselines on public benchmarks.

Demonstrates effectiveness through online A/B tests.

Abstract

Auto-bidding systems aim to maximize marketing value while satisfying strict efficiency constraints such as Target Cost-Per-Action (CPA). Although Decision Transformers provide powerful sequence modeling capabilities, applying them to this constrained setting encounters two challenges: 1) standard Return-to-Go conditioning causes state aliasing by neglecting the cost dimension, preventing precise resource pacing; and 2) standard regression forces the policy to mimic average historical behaviors, thereby limiting the capacity to optimize performance toward the constraint boundary. To address these challenges, we propose PRO-Bid, a constraint-aware generative auto-bidding framework based on two synergistic mechanisms: 1) Constraint-Decoupled Pareto Representation (CDPR) decomposes global constraints into recursive cost and value contexts to restore resource perception, while reweighting trajectories based on the Pareto frontier to focus on high-efficiency data; and 2) Counterfactual Regret Optimization (CRO) facilitates active improvement by utilizing a global outcome predictor to identify superior counterfactual actions. By treating these high-utility outcomes as weighted regression targets, the model transcends historical averages to approach the optimal constraint boundary. Extensive experiments on two public benchmarks and online A/B tests demonstrate that PRO-Bid achieves superior constraint satisfaction and value acquisition compared to state-of-the-art baselines.