pith. sign in

For most experiments, we use a batch size of 128, a learning rate of 10−4, and an exponential moving aver- age (EMA) over model parameters with a rate of 0.9999

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.LG 1

years

2021 1

verdicts

ACCEPT 1

representative citing papers

Improved Denoising Diffusion Probabilistic Models

cs.LG · 2021-02-18 · accept · novelty 7.0

Targeted tweaks to DDPMs produce competitive likelihoods and high-quality samples, with learned reverse variances enabling 10x faster sampling and predictable scaling with compute.

citing papers explorer

Showing 1 of 1 citing paper.

  • Improved Denoising Diffusion Probabilistic Models cs.LG · 2021-02-18 · accept · none · ref 12

    Targeted tweaks to DDPMs produce competitive likelihoods and high-quality samples, with learned reverse variances enabling 10x faster sampling and predictable scaling with compute.