论文标题
共轭自然选择
Conjugate Natural Selection
论文作者
论文摘要
我们证明，Fisher-Rao天然梯度下降（FR-NGD）最佳地近似连续的时间复制器方程（进化动力学的基本模型），并称此对应关系为“共轭自然选择”。这种对应有望在连续或高维假设空间上进行进化计算的替代方法。作为一种特殊情况，当假设基于预测实际观察结果竞争时，FR-NGD还提供了连续贝叶斯推断的最佳近似。在这种情况下，该方法避免需要计算先前的概率。我们证明了我们在非凸优化问题和随机变化参数的随机过程中的系统标识任务。
We prove that Fisher-Rao natural gradient descent (FR-NGD) optimally approximates the continuous time replicator equation (an essential model of evolutionary dynamics), and term this correspondence "conjugate natural selection". This correspondence promises alternative approaches for evolutionary computation over continuous or high-dimensional hypothesis spaces. As a special case, FR-NGD also provides the optimal approximation of continuous Bayesian inference when hypotheses compete on the basis of predicting actual observations. In this case, the method avoids the need to compute prior probabilities. We demonstrate our findings on a non-convex optimization problem and a system identification task for a stochastic process with time-varying parameters.
