179 citations · 505 across the 46 of their papers we have counts for
5 papers · 1 filter
HetSeq: Distributed GPU Training on Heterogeneous Infrastructure
Yifan Ding, Nicholas Botzer, Tim Weninger
Modern deep learning systems like PyTorch and Tensorflow are able to train enormous models with billions (or trillions) of parameters on a distributed infrastructure. These systems…
The Infinity Mirror Test for Graph Models
Satyaki Sikdar, Daniel Gonzalez Cedre, Trenton W. Ford +1
Graph models, like other machine learning models, have implicit and explicit biases built-in, which often impact performance in nontrivial ways. The model's faithfulness is often m…
Joint Subgraph-to-Subgraph Transitions -- Generalizing Triadic Closure for Powerful and Interpretable Graph Modeling
Justus Hibshman, Daniel Gonzalez Cedre, Satyaki Sikdar +1
We generalize triadic closure, along with previous generalizations of triadic closure, under an intuitive umbrella generalization: the Subgraph-to-Subgraph Transition (SST). We pre…
Library Adoption Dynamics in Software Teams
Pamela Bilo Thomas, Rachel Krohn, Tim Weninger
When a group of people strives to understand new information, struggle ensues as various ideas compete for attention. Steep learning curves are surmounted as teams learn together.…
Automatic Discovery of Political Meme Genres with Diverse Appearances
William Theisen, Joel Brogan, Pamela Bilo Thomas +4
Forms of human communication are not static -- we expect some evolution in the way information is conveyed over time because of advances in technology. One example of this phenomen…