Alert button
Picture for Badih Ghazi

Badih Ghazi

Alert button

Algorithms with More Granular Differential Privacy Guarantees

Sep 08, 2022
Badih Ghazi, Ravi Kumar, Pasin Manurangsi, Thomas Steinke

Figure 1 for Algorithms with More Granular Differential Privacy Guarantees
Viaarxiv icon

Faster Privacy Accounting via Evolving Discretization

Jul 10, 2022
Badih Ghazi, Pritish Kamath, Ravi Kumar, Pasin Manurangsi

Figure 1 for Faster Privacy Accounting via Evolving Discretization
Figure 2 for Faster Privacy Accounting via Evolving Discretization
Figure 3 for Faster Privacy Accounting via Evolving Discretization
Viaarxiv icon

Connect the Dots: Tighter Discrete Approximations of Privacy Loss Distributions

Jul 10, 2022
Vadym Doroshenko, Badih Ghazi, Pritish Kamath, Ravi Kumar, Pasin Manurangsi

Figure 1 for Connect the Dots: Tighter Discrete Approximations of Privacy Loss Distributions
Figure 2 for Connect the Dots: Tighter Discrete Approximations of Privacy Loss Distributions
Figure 3 for Connect the Dots: Tighter Discrete Approximations of Privacy Loss Distributions
Figure 4 for Connect the Dots: Tighter Discrete Approximations of Privacy Loss Distributions
Viaarxiv icon

User-Level Private Learning via Correlated Sampling

Oct 21, 2021
Badih Ghazi, Ravi Kumar, Pasin Manurangsi

Figure 1 for User-Level Private Learning via Correlated Sampling
Viaarxiv icon

Large-Scale Differentially Private BERT

Aug 03, 2021
Rohan Anil, Badih Ghazi, Vineet Gupta, Ravi Kumar, Pasin Manurangsi

Figure 1 for Large-Scale Differentially Private BERT
Figure 2 for Large-Scale Differentially Private BERT
Figure 3 for Large-Scale Differentially Private BERT
Figure 4 for Large-Scale Differentially Private BERT
Viaarxiv icon

Locally Private k-Means in One Round

May 15, 2021
Alisa Chang, Badih Ghazi, Ravi Kumar, Pasin Manurangsi

Figure 1 for Locally Private k-Means in One Round
Figure 2 for Locally Private k-Means in One Round
Viaarxiv icon

On Deep Learning with Label Differential Privacy

Feb 11, 2021
Badih Ghazi, Noah Golowich, Ravi Kumar, Pasin Manurangsi, Chiyuan Zhang

Figure 1 for On Deep Learning with Label Differential Privacy
Figure 2 for On Deep Learning with Label Differential Privacy
Figure 3 for On Deep Learning with Label Differential Privacy
Figure 4 for On Deep Learning with Label Differential Privacy
Viaarxiv icon

On Avoiding the Union Bound When Answering Multiple Differentially Private Queries

Dec 16, 2020
Badih Ghazi, Ravi Kumar, Pasin Manurangsi

Viaarxiv icon

Sample-efficient proper PAC learning with approximate differential privacy

Dec 07, 2020
Badih Ghazi, Noah Golowich, Ravi Kumar, Pasin Manurangsi

Figure 1 for Sample-efficient proper PAC learning with approximate differential privacy
Viaarxiv icon