Picture for Rudrajit Das

Rudrajit Das

Retraining with Predicted Hard Labels Provably Increases Model Accuracy

Add code
Jun 17, 2024
Viaarxiv icon

Towards Quantifying the Preconditioning Effect of Adam

Add code
Feb 11, 2024
Viaarxiv icon

Understanding the Training Speedup from Sampling with Approximate Losses

Add code
Feb 10, 2024
Viaarxiv icon

Understanding Self-Distillation in the Presence of Label Noise

Add code
Jan 30, 2023
Viaarxiv icon

Beyond Uniform Lipschitz Condition in Differentially Private Optimization

Add code
Jun 21, 2022
Figure 1 for Beyond Uniform Lipschitz Condition in Differentially Private Optimization
Figure 2 for Beyond Uniform Lipschitz Condition in Differentially Private Optimization
Figure 3 for Beyond Uniform Lipschitz Condition in Differentially Private Optimization
Figure 4 for Beyond Uniform Lipschitz Condition in Differentially Private Optimization
Viaarxiv icon

On the Unreasonable Effectiveness of Federated Averaging with Heterogeneous Data

Add code
Jun 09, 2022
Figure 1 for On the Unreasonable Effectiveness of Federated Averaging with Heterogeneous Data
Figure 2 for On the Unreasonable Effectiveness of Federated Averaging with Heterogeneous Data
Figure 3 for On the Unreasonable Effectiveness of Federated Averaging with Heterogeneous Data
Viaarxiv icon

DISCO : efficient unsupervised decoding for discrete natural language problems via convex relaxation

Add code
Jul 13, 2021
Figure 1 for DISCO : efficient unsupervised decoding for discrete natural language problems via convex relaxation
Figure 2 for DISCO : efficient unsupervised decoding for discrete natural language problems via convex relaxation
Viaarxiv icon

DP-NormFedAvg: Normalizing Client Updates for Privacy-Preserving Federated Learning

Add code
Jun 13, 2021
Figure 1 for DP-NormFedAvg: Normalizing Client Updates for Privacy-Preserving Federated Learning
Figure 2 for DP-NormFedAvg: Normalizing Client Updates for Privacy-Preserving Federated Learning
Figure 3 for DP-NormFedAvg: Normalizing Client Updates for Privacy-Preserving Federated Learning
Figure 4 for DP-NormFedAvg: Normalizing Client Updates for Privacy-Preserving Federated Learning
Viaarxiv icon

Improved Convergence Rates for Non-Convex Federated Learning with Compression

Add code
Dec 12, 2020
Figure 1 for Improved Convergence Rates for Non-Convex Federated Learning with Compression
Viaarxiv icon

On the Benefits of Multiple Gossip Steps in Communication-Constrained Decentralized Optimization

Add code
Nov 20, 2020
Figure 1 for On the Benefits of Multiple Gossip Steps in Communication-Constrained Decentralized Optimization
Figure 2 for On the Benefits of Multiple Gossip Steps in Communication-Constrained Decentralized Optimization
Figure 3 for On the Benefits of Multiple Gossip Steps in Communication-Constrained Decentralized Optimization
Figure 4 for On the Benefits of Multiple Gossip Steps in Communication-Constrained Decentralized Optimization
Viaarxiv icon