arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2201.02890cs.LGcs.NIstat.ML
收藏

用于在线学习的带预测的Lazy Lagrangians

Lazy Lagrangians with Predictions for Online Learning

Trinity College Dublin(都柏林圣三一学院);Delft University of Technology(代尔夫特理工大学)

Daron Anderson, George Iosifidis, Douglas J. Leith

摘要:

我们考虑一般的在线凸优化问题与时变的附加约束下的预测下一个成本和约束函数。将Follow-The-Regularized-Leader迭代与预测自适应动态步长相结合,设计了一种新的原始-对偶算法。该算法实现了$\mathcal O(T^{\frac{3-β}{4}})$遗憾和$\mathcal O(T^{\frac{1+β}{2}})$约束违反边界,可通过参数$β\!\in\![1/2,1)$进行调整,并具有随预测质量收缩的常数因子,最终实现$\mathcal O(1)$遗憾完美的预测。我们的工作扩展了FTRL框架的约束OCO设置和优于各自的国家的最先进的贪婪为基础的解决方案,而不施加条件的预测质量,成本函数或几何约束,超越凸性。

英文摘要:

We consider the general problem of online convex optimization with time-varying additive constraints in the presence of predictions for the next cost and constraint functions. A novel primal-dual algorithm is designed by combining a Follow-The-Regularized-Leader iteration with prediction-adaptive dynamic steps. The algorithm achieves $\mathcal O(T^{\frac{3-β}{4}})$ regret and $\mathcal O(T^{\frac{1+β}{2}})$ constraint violation bounds that are tunable via parameter $β\!\in\![1/2,1)$ and have constant factors that shrink with the predictions quality, achieving eventually $\mathcal O(1)$ regret for perfect predictions. Our work extends the FTRL framework for this constrained OCO setting and outperforms the respective state-of-the-art greedy-based solutions, without imposing conditions on the quality of predictions, the cost functions or the geometry of constraints, beyond convexity.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑