Vaibhav Kumar Dixit

At this line, https://github.com/wildart/Evolutionary.jl/blob/cf3f2fbf5e1be13d6f5db23eabe5dbfab4d5b688/src/es.jl#L100, where the `recombination` function is invoked in the `ES` implementation, because `rng` is passed as a kwarg unconditionally, using the default https://github.com/wildart/Evolutionary.jl/blob/cf3f2fbf5e1be13d6f5db23eabe5dbfab4d5b688/src/es.jl#L30, causes an error. https://github.com/SciML/Optimization.jl/issues/467

[WIP] Add Manopt.jl wrapper

6

Aimed at moving the already substantial work done in #437 forward ## Checklist - [ ] Appropriate tests were added - [ ] Any code changes were done in a...

## Checklist - [ ] Appropriate tests were added - [ ] Any code changes were done in a way that does not break public API - [ ] All...

As part of docs updates there should be an addition of how to use this

OptimizationPolyAlgorithms can become the primary choice by adding these methods, pretty low hanging fruit

**Describe the bug 🐞** Currently the error thrown when a user has not used `OptimizationFunction` is something like `grad not found`. The way to do this would be define a...

[WIP] Make cache mutable struct

1

## Checklist - [ ] Appropriate tests were added - [ ] Any code changes were done in a way that does not break public API - [ ] All...

Use larger step size in ADAM in polyalgorithm

1

Using very small stepsize doesn't make sense since the point is to more effectively explore the parameter space