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

Add to Chrome

Add to Firefox

Picture for Praneeth Vepakomma

Praneeth Vepakomma

Differentially Private CutMix for Split Learning with Vision Transformer

Oct 28, 2022
Seungeun Oh, Jihong Park, Sihun Baek, Hyelin Nam, Praneeth Vepakomma, Ramesh Raskar, Mehdi Bennis, Seong-Lyun Kim

Figure 1 for Differentially Private CutMix for Split Learning with Vision Transformer

Figure 2 for Differentially Private CutMix for Split Learning with Vision Transformer

Figure 3 for Differentially Private CutMix for Split Learning with Vision Transformer

Figure 4 for Differentially Private CutMix for Split Learning with Vision Transformer

* to be presented at the 36nd Conference on Neural Information Processing Systems (NeurIPS 2022), First Workshop on Interpolation Regularizers and Beyond (INTERPOLATE), New Orleans, United States

Access Paper or Ask Questions

Differentially Private Fréchet Mean on the Manifold of Symmetric Positive Definite (SPD) Matrices

Aug 08, 2022
Saiteja Utpala, Praneeth Vepakomma, Nina Miolane

Figure 1 for Differentially Private Fréchet Mean on the Manifold of Symmetric Positive Definite (SPD) Matrices

Figure 2 for Differentially Private Fréchet Mean on the Manifold of Symmetric Positive Definite (SPD) Matrices

Figure 3 for Differentially Private Fréchet Mean on the Manifold of Symmetric Positive Definite (SPD) Matrices

Figure 4 for Differentially Private Fréchet Mean on the Manifold of Symmetric Positive Definite (SPD) Matrices

* 28 pages

Access Paper or Ask Questions

Visual Transformer Meets CutMix for Improved Accuracy, Communication Efficiency, and Data Privacy in Split Learning

Jul 01, 2022
Sihun Baek, Jihong Park, Praneeth Vepakomma, Ramesh Raskar, Mehdi Bennis, Seong-Lyun Kim

Figure 1 for Visual Transformer Meets CutMix for Improved Accuracy, Communication Efficiency, and Data Privacy in Split Learning

Figure 2 for Visual Transformer Meets CutMix for Improved Accuracy, Communication Efficiency, and Data Privacy in Split Learning

Figure 3 for Visual Transformer Meets CutMix for Improved Accuracy, Communication Efficiency, and Data Privacy in Split Learning

Figure 4 for Visual Transformer Meets CutMix for Improved Accuracy, Communication Efficiency, and Data Privacy in Split Learning

* won the Best Student Paper Award at International Workshop on Trustworthy Federated Learning in Conjunction with IJCAI 2022 (FL-IJCAI'22), Vienna, Austria

Access Paper or Ask Questions

Server-Side Local Gradient Averaging and Learning Rate Acceleration for Scalable Split Learning

Dec 11, 2021
Shraman Pal, Mansi Uniyal, Jihong Park, Praneeth Vepakomma, Ramesh Raskar, Mehdi Bennis, Moongu Jeon, Jinho Choi

Figure 1 for Server-Side Local Gradient Averaging and Learning Rate Acceleration for Scalable Split Learning

Figure 2 for Server-Side Local Gradient Averaging and Learning Rate Acceleration for Scalable Split Learning

Figure 3 for Server-Side Local Gradient Averaging and Learning Rate Acceleration for Scalable Split Learning

Figure 4 for Server-Side Local Gradient Averaging and Learning Rate Acceleration for Scalable Split Learning

* 9 pages, 3 figures, 6 tables

Access Paper or Ask Questions

Private measurement of nonlinear correlations between data hosted across multiple parties

Nov 08, 2021
Praneeth Vepakomma, Subha Nawer Pushpita, Ramesh Raskar

Figure 1 for Private measurement of nonlinear correlations between data hosted across multiple parties

Figure 2 for Private measurement of nonlinear correlations between data hosted across multiple parties

Figure 3 for Private measurement of nonlinear correlations between data hosted across multiple parties

Figure 4 for Private measurement of nonlinear correlations between data hosted across multiple parties

Access Paper or Ask Questions

Parallel Quasi-concave set optimization: A new frontier that scales without needing submodularity

Aug 19, 2021
Praneeth Vepakomma, Yulia Kempner, Ramesh Raskar

Figure 1 for Parallel Quasi-concave set optimization: A new frontier that scales without needing submodularity

Figure 2 for Parallel Quasi-concave set optimization: A new frontier that scales without needing submodularity

Figure 3 for Parallel Quasi-concave set optimization: A new frontier that scales without needing submodularity

Figure 4 for Parallel Quasi-concave set optimization: A new frontier that scales without needing submodularity

* SubSetML: Subset Selection in Machine Learning: From Theory to Practice

Access Paper or Ask Questions