Ameya Daigavane

We have some recent work that enables node-level privacy for training GNNs! This was accepted as an Oral at PAIR2Struct, ICLR 2022: "Node-Level Differentially Private Graph Neural Networks" Ameya Daigavane,...

We believe our interactive article published at Distill would be a valuable addition to the Survey section: "Understanding Convolutions on Graphs", Distill, 2021. Ameya Daigavane and Balaraman Ravindran and Gaurav...

Input own similarity matrix?

1

Thanks for the implementation. We have a similarity matrix W, and we were wondering if we could use SpectralNet to learn embeddings (in the spectral basis of the associated Laplacian...

Fixes Gaussian PSF installation errors.

1

Resolves #1.

ODE Solver with Constraints?

2

Hi, I was wondering if Diffrax supported adding a norm constraint at the boundary: $$|y(1)| = \text{constant}$$ for the neural ODE: $$\frac{dy}{dt} = f(t, y), \ y(0) = x$$ Any...

How to use reversible integrators?

1

I wanted a time-reversible integrator, and according to the documentation SemiImplicitEuler() should work. However, I ran the following: ```python def f(t, p, args): return -p def g(t, q, args): return...

Is there a comparison with OpenBabel's bond order assignment algorithm somewhere? Would be super helpful :)


This looks great! Is there a way to highlight certain atoms/bonds on clicking them?

The imports in the current HMM example seem invalid, maybe because the directory structure of the package has changed. ``` AttributeError: 'module' object has no attribute 'distributions' ``` I just...