Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Gintare Karolina Dziugaite

Gintare Karolina Dziugaite

Element AI

Pruning's Effect on Generalization Through the Lens of Training and Regularization

Oct 25, 2022
Tian Jin, Michael Carbin, Daniel M. Roy, Jonathan Frankle, Gintare Karolina Dziugaite

Figure 1 for Pruning's Effect on Generalization Through the Lens of Training and Regularization

Figure 2 for Pruning's Effect on Generalization Through the Lens of Training and Regularization

Figure 3 for Pruning's Effect on Generalization Through the Lens of Training and Regularization

Figure 4 for Pruning's Effect on Generalization Through the Lens of Training and Regularization

* Advances in Neural Information Processing Systems 2022
* 49 pages, 20 figures

Access Paper or Ask Questions

Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?

Oct 06, 2022
Mansheej Paul, Feng Chen, Brett W. Larsen, Jonathan Frankle, Surya Ganguli, Gintare Karolina Dziugaite

Figure 1 for Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?

Figure 2 for Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?

Figure 3 for Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?

Figure 4 for Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?

* The first three authors contributed equally

Access Paper or Ask Questions

Understanding Generalization via Leave-One-Out Conditional Mutual Information

Jun 29, 2022
Mahdi Haghifam, Shay Moran, Daniel M. Roy, Gintare Karolina Dziugaite

Figure 1 for Understanding Generalization via Leave-One-Out Conditional Mutual Information

Figure 2 for Understanding Generalization via Leave-One-Out Conditional Mutual Information

* 18 pages

Access Paper or Ask Questions

Lottery Tickets on a Data Diet: Finding Initializations with Sparse Trainable Networks

Jun 02, 2022
Mansheej Paul, Brett W. Larsen, Surya Ganguli, Jonathan Frankle, Gintare Karolina Dziugaite

Figure 1 for Lottery Tickets on a Data Diet: Finding Initializations with Sparse Trainable Networks

Figure 2 for Lottery Tickets on a Data Diet: Finding Initializations with Sparse Trainable Networks

Figure 3 for Lottery Tickets on a Data Diet: Finding Initializations with Sparse Trainable Networks

Figure 4 for Lottery Tickets on a Data Diet: Finding Initializations with Sparse Trainable Networks

* The first two authors contributed equally

Access Paper or Ask Questions

Towards a Unified Information-Theoretic Framework for Generalization

Nov 17, 2021
Mahdi Haghifam, Gintare Karolina Dziugaite, Shay Moran, Daniel M. Roy

Figure 1 for Towards a Unified Information-Theoretic Framework for Generalization

* 22 Pages, NeurIPS 2021, This submission subsumes [arXiv:2011.02970] ("On the Information Complexity of Proper Learners for VC Classes in the Realizable Case")

Access Paper or Ask Questions

Probabilistic fine-tuning of pruning masks and PAC-Bayes self-bounded learning

Oct 22, 2021
Soufiane Hayou, Bobby He, Gintare Karolina Dziugaite

Figure 1 for Probabilistic fine-tuning of pruning masks and PAC-Bayes self-bounded learning

Figure 2 for Probabilistic fine-tuning of pruning masks and PAC-Bayes self-bounded learning

Figure 3 for Probabilistic fine-tuning of pruning masks and PAC-Bayes self-bounded learning

Figure 4 for Probabilistic fine-tuning of pruning masks and PAC-Bayes self-bounded learning

* 34 pages, 10 figures

Access Paper or Ask Questions

Deep Learning on a Data Diet: Finding Important Examples Early in Training

Jul 15, 2021
Mansheej Paul, Surya Ganguli, Gintare Karolina Dziugaite

Figure 1 for Deep Learning on a Data Diet: Finding Important Examples Early in Training

Figure 2 for Deep Learning on a Data Diet: Finding Important Examples Early in Training

Figure 3 for Deep Learning on a Data Diet: Finding Important Examples Early in Training

Figure 4 for Deep Learning on a Data Diet: Finding Important Examples Early in Training

* 18 pages, 16 figures

Access Paper or Ask Questions

NeurIPS 2020 Competition: Predicting Generalization in Deep Learning

Dec 14, 2020
Yiding Jiang, Pierre Foret, Scott Yak, Daniel M. Roy, Hossein Mobahi, Gintare Karolina Dziugaite, Samy Bengio, Suriya Gunasekar, Isabelle Guyon, Behnam Neyshabur

Figure 1 for NeurIPS 2020 Competition: Predicting Generalization in Deep Learning

* 20 pages, 2 figures. Accepted for NeurIPS 2020 Competitions Track. Lead organizer: Yiding Jiang

Access Paper or Ask Questions