-
It would be interesting to start exploring asynchronous algorithms within this project using the dask.distributed API. Because this API is somewhat different it might be wise to start with something …
-
This issue keeps track of the support for constraints in GPFlowOpt.
Support for expensive constraints will be initially added using the Probability of Feasbility (PoF). With the acquisition functio…
-
## Keyword: sgd
### Doubly Stochastic Models: Learning with Unbiased Label Noises and Inference Stability
- **Authors:** Authors: Haoyi Xiong, Xuhong Li, Boyang Yu, Zhanxing Zhu, Dongrui Wu, Dejin…
-
@akaptano , @znicolaou
When I proposed to combine the derivative and sparse regression step into a single optimization problem, it was clear that I couldn't use the existing SINDy class without so…
-
Since many contractors are based on `opt_einsum` it might be useful to add support for its [path reuse functionality](https://optimized-einsum.readthedocs.io/en/latest/reusing_paths.html). This would …
-
Hello,
Feel free to close this if this is out of scope for this project.
I started working on a C interface for OSI: https://github.com/lovasoa/c-osi
What would you think about having an offi…
-
I would like to be able to control the finite difference scheme used, i.e. forward, backward or central. Depending upon the PDE, we normally use a custom scheme, e.g. advection --> backwards, diffusio…
-
I think `NamedTuple`s are allowed as `Active`s, but currently, we get (may be a known problem):
```julia
julia> autodiff(x -> x.a * x.b, Active((a = 2, b = 3)))
((a = 0, b = 0),)
```
-
## Description
When seeding STOMP with a valid trajectory (from RRT), STOMP fails to find a solution by reaching the max number of iterations.
 is a non-linear least squares package written …