James Arten's user avatar
James Arten's user avatar
James Arten's user avatar
James Arten
  • Member for 2 years, 6 months
  • Last seen more than a month ago
  • United States
6 votes
2 answers
771 views

What is the exact role of model $p_\theta$ in diffusion models for the reverse process?

3 votes
1 answer
1k views

What is multi-head attention doing mathematically, and how is it different from self-attention?

2 votes
1 answer
69 views

Is there a way to inject linear constrains during GAN training?

2 votes
1 answer
1k views

Could it make any sense to choose a larger dimension for the latent space of the VAE with respect to the original input?

2 votes
0 answers
211 views

Are spectral approaches to Graph Neural Networks still considered?

2 votes
1 answer
373 views

Why do we use a linear interpolation of fake and real data to penalize the gradient of discriminator in WGAN-GP

2 votes
2 answers
114 views

Definition of negatives in NT-Xent loss

1 vote
0 answers
337 views

Out of distribution detection (OOD) in the context of regression problems

0 votes
0 answers
59 views

Understanding relation between parameter sharing and Message Passing in Graph Neural Networks

0 votes
0 answers
12 views

Explicit representation learning task benefits than simply considering last nerual network layer

0 votes
0 answers
25 views

How does the latent space learned by a neural network look like, in the regression case?

0 votes
0 answers
34 views

Why the positive example is included in the denominator of NT-Xent loss?

0 votes
0 answers
1k views

How to encode time embeddings in a diffusion model for 1D vectors?

0 votes
1 answer
219 views

Parametric vs Non-parametric generative models

0 votes
1 answer
156 views

What is intended for 1x1 convolution for input images?

0 votes
1 answer
88 views

Are positional embeddings computed during or before training?

0 votes
0 answers
128 views

In CVAE's objective function, why do both terms condition on $\textbf{c}$?