Alert button
Picture for Daniel M. Roy

Daniel M. Roy

Alert button

Fast-rate PAC-Bayes Generalization Bounds via Shifted Rademacher Processes

Aug 20, 2019
Jun Yang, Shengyang Sun, Daniel M. Roy

Viaarxiv icon

Black-box constructions for exchangeable sequences of random multisets

Aug 17, 2019
Creighton Heaukulani, Daniel M. Roy

Viaarxiv icon

NUQSGD: Improved Communication Efficiency for Data-parallel SGD via Nonuniform Quantization

Aug 16, 2019
Ali Ramezani-Kebrya, Fartash Faghri, Daniel M. Roy

Figure 1 for NUQSGD: Improved Communication Efficiency for Data-parallel SGD via Nonuniform Quantization
Figure 2 for NUQSGD: Improved Communication Efficiency for Data-parallel SGD via Nonuniform Quantization
Figure 3 for NUQSGD: Improved Communication Efficiency for Data-parallel SGD via Nonuniform Quantization
Figure 4 for NUQSGD: Improved Communication Efficiency for Data-parallel SGD via Nonuniform Quantization
Viaarxiv icon

The Lottery Ticket Hypothesis at Scale

Mar 05, 2019
Jonathan Frankle, Gintare Karolina Dziugaite, Daniel M. Roy, Michael Carbin

Figure 1 for The Lottery Ticket Hypothesis at Scale
Figure 2 for The Lottery Ticket Hypothesis at Scale
Figure 3 for The Lottery Ticket Hypothesis at Scale
Figure 4 for The Lottery Ticket Hypothesis at Scale
Viaarxiv icon

Entropy-SGD optimizes the prior of a PAC-Bayes bound: Generalization properties of Entropy-SGD and data-dependent priors

Mar 03, 2018
Gintare Karolina Dziugaite, Daniel M. Roy

Figure 1 for Entropy-SGD optimizes the prior of a PAC-Bayes bound: Generalization properties of Entropy-SGD and data-dependent priors
Figure 2 for Entropy-SGD optimizes the prior of a PAC-Bayes bound: Generalization properties of Entropy-SGD and data-dependent priors
Figure 3 for Entropy-SGD optimizes the prior of a PAC-Bayes bound: Generalization properties of Entropy-SGD and data-dependent priors
Figure 4 for Entropy-SGD optimizes the prior of a PAC-Bayes bound: Generalization properties of Entropy-SGD and data-dependent priors
Viaarxiv icon

Data-dependent PAC-Bayes priors via differential privacy

Feb 26, 2018
Gintare Karolina Dziugaite, Daniel M. Roy

Figure 1 for Data-dependent PAC-Bayes priors via differential privacy
Figure 2 for Data-dependent PAC-Bayes priors via differential privacy
Viaarxiv icon

Exchangeable modelling of relational data: checking sparsity, train-test splitting, and sparse exchangeable Poisson matrix factorization

Dec 06, 2017
Victor Veitch, Ekansh Sharma, Zacharie Naulet, Daniel M. Roy

Figure 1 for Exchangeable modelling of relational data: checking sparsity, train-test splitting, and sparse exchangeable Poisson matrix factorization
Figure 2 for Exchangeable modelling of relational data: checking sparsity, train-test splitting, and sparse exchangeable Poisson matrix factorization
Figure 3 for Exchangeable modelling of relational data: checking sparsity, train-test splitting, and sparse exchangeable Poisson matrix factorization
Figure 4 for Exchangeable modelling of relational data: checking sparsity, train-test splitting, and sparse exchangeable Poisson matrix factorization
Viaarxiv icon

Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Data

Oct 19, 2017
Gintare Karolina Dziugaite, Daniel M. Roy

Figure 1 for Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Data
Figure 2 for Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Data
Viaarxiv icon

A study of the effect of JPG compression on adversarial images

Aug 02, 2016
Gintare Karolina Dziugaite, Zoubin Ghahramani, Daniel M. Roy

Figure 1 for A study of the effect of JPG compression on adversarial images
Figure 2 for A study of the effect of JPG compression on adversarial images
Figure 3 for A study of the effect of JPG compression on adversarial images
Figure 4 for A study of the effect of JPG compression on adversarial images
Viaarxiv icon

A characterization of product-form exchangeable feature probability functions

Jul 07, 2016
Marco Battiston, Stefano Favaro, Daniel M. Roy, Yee Whye Teh

Viaarxiv icon