论文标题
构图亚当：自适应成分求解器
Compositional ADAM: An Adaptive Compositional Solver
论文作者
论文摘要
在本文中，我们提出了C-ADAM，这是涉及期望值非线性功能嵌套的组成问题的第一个自适应求解器。我们证明，C-ADAM以$ \ Mathcal {O}（δ^{ - 2.25}）$收敛到固定点，而$δ$作为精度参数。此外，我们通过首次桥接模型不稳定元学习（MAML）和组成优化，证明了结果的重要性，显示了最快的已知速率，以供深层网络适应至今。最后，我们通过投资组合优化和元学习的一组实验验证了我们的发现。与标准求解器和组成求解器相比，我们的结果表现出显着的样品复杂性降低。
In this paper, we present C-ADAM, the first adaptive solver for compositional problems involving a non-linear functional nesting of expected values. We proof that C-ADAM converges to a stationary point in $\mathcal{O}(δ^{-2.25})$ with $δ$ being a precision parameter. Moreover, we demonstrate the importance of our results by bridging, for the first time, model-agnostic meta-learning (MAML) and compositional optimisation showing fastest known rates for deep network adaptation to-date. Finally, we validate our findings in a set of experiments from portfolio optimisation and meta-learning. Our results manifest significant sample complexity reductions compared to both standard and compositional solvers.
