Kamer Ali Yuksel

Question about the method

2

What prevents the model from returning back to the same local minima while having cyclic learning rates or warm restarts? Would it make sense to regularize the model to make...

L2 Regularization for Sparsity?

5

L2 regularization as explained in your documentation penalizes large weights and hence would make your portfolio more diverse. I don't think that it has a positive effect on eliminating negligible...

[Feature Request] Adaptive Gradient Clipping

2

It would be cool to use AGC from the following paper with any optimizer in this package: https://arxiv.org/abs/2102.06171

Using Approximate k-Nearest Neighbourhood Queries (Support for Integrating Popular ANN Libraries and Using as a Proximity Backend)

2

Would it be possible to integrate and use any of the approximate k-NN libraries such as HNSWLIB, FAISS (Facebook), ANNOY (Spotify), or SCANN (Google) with many proximity-based methods, which are...

Is that functionality supported somehow?

Dear Bayesians, I am working on a project where I sample a set of n-dimensional points from a Gaussian distribution (of learnt parameters) as follows and then evaluate those points...

Would it be possible for you to also provide a 1D version of the layer to be used for time-series classification / regression? Thanks!

[Feature Request] Optimal Portfolio Allocation via Independent Component Analysis

2

IC-variance-parity portfolio: Factor-Risk-Parity Portfolio based on Maximally Independent Factors (via ICA) [Optimal_Portfolio_ICA.pdf](https://github.com/jankrepl/deepdow/files/7101319/Optimal_Portfolio_ICA.pdf)

DropBlock1D

3

Can you please also include DropBlock1D implementation to use it for time-series? Thank you very much.