用自编码器快速修正不可行预测，提升复杂约束下的系统效率
Improving Feasibility via Fast Autoencoder-Based Projections
- 训练对抗性自编码器学习可行集的凸潜空间，实现快速投影
- 在多种非凸约束任务中，以低计算成本100%满足约束条件
- 适合需要实时可行性保障的强化学习与优化系统
在真实学习与控制系统中，强制执行复杂（如非凸）操作约束是一个关键挑战。现有方法难以高效处理一般类约束。为此，我们提出一种新型数据驱动的摊销方法，利用训练好的自编码器作为近似投影器，对不可行预测进行快速修正。具体而言，通过对抗性目标训练自编码器，学习可行集的结构化凸潜表示，从而在解码前将神经网络输出的潜表示投影到简单凸形状，快速恢复至原始可行集。我们在包含挑战性非凸约束的多样化约束优化与强化学习问题上测试该方法。结果表明，该方法以极低计算成本有效强制执行约束，为传统求解器依赖的昂贵可行性修正技术提供了实用替代方案。
原文摘要 · Abstract (English)
Enforcing complex (e.g., nonconvex) operational constraints is a critical challenge in real-world learning and control systems. However, existing methods struggle to efficiently enforce general classes of constraints. To address this, we propose a novel data-driven amortized approach that uses a trained autoencoder as an approximate projector to provide fast corrections to infeasible predictions. Specifically, we train an autoencoder using an adversarial objective to learn a structured, convex latent representation of the feasible set. This enables rapid correction of neural network outputs by projecting their associated latent representations onto a simple convex shape before decoding into the original feasible set. We test our approach on a diverse suite of constrained optimization and reinforcement learning problems with challenging nonconvex constraints. Results show that our method effectively enforces constraints at a low computational cost, offering a practical alternative to expensive feasibility correction techniques based on traditional solvers.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。