Randomization Approach

Randomization techniques are increasingly employed to improve the robustness and performance of various machine learning models and algorithms. Current research focuses on optimizing randomization strategies within deep neural networks, exploring their impact on generalization, overfitting, and Sim2Real transfer in robotics, and applying them to address challenges like copyright infringement and optical system calibration. These methods show promise in enhancing model efficiency and performance across diverse applications, particularly where data scarcity or domain mismatch is a significant hurdle. The development and refinement of these techniques are crucial for advancing the reliability and applicability of machine learning in various fields.

Papers

August 21, 2024

Randomization Techniques to Mitigate the Risk of Copyright Infringement
Wei-Ning Chen, Peter Kairouz, Sewoong Oh, Zheng Xu
Differential Privacy High Similarity High Quality Risk Description Copyright Protection Copyright Violation Randomization Approach

April 5, 2024

Rolling the dice for better deep learning performance: A study of randomness techniques in deep neural networks
Mohammed Ghaith Altarabichi, Sławomir Nowaczyk, Sepideh Pashami, Peyman Sheikholharam Mashhadi, Julia Handl
Neural Network Deep Neural Network Study Feature Randomization Program Improved Box Dice Randomness Factor Initial Weight Randomization Approach

June 13, 2022

Analysis of Randomization Effects on Sim2Real Transfer in Reinforcement Learning for Robotic Manipulation Tasks
Josip Josifovski, Mohammadhossein Malmir, Noah Klarmann, Bare Luka Žagar, Nicolás Navarro-Guerrero, Alois Knoll
Reinforcement Learning General Analysis Manipulation Task Randomization Program Randomized Experiment Sim2Real Transfer Randomization Approach

February 25, 2022

Domain Adaptation: the Key Enabler of Neural Network Equalizers in Coherent Optical Systems
Pedro J. Freire, Bernhard Spinnler, Daniel Abode, Jaroslaw E. Prilepsky, Abdallah A. I. Ali, Nelson Costa, Wolfgang Schairer, Antonio Napoli, Andrew D. Ellis, Sergei K. Turitsyn
Domain Adaptation Synthetic Data Transmission Data Neural Network Based Equalizer Coherent Optical Randomization Approach

Randomization Approach

Papers

Randomization Techniques to Mitigate the Risk of Copyright Infringement

Rolling the dice for better deep learning performance: A study of randomness techniques in deep neural networks

Analysis of Randomization Effects on Sim2Real Transfer in Reinforcement Learning for Robotic Manipulation Tasks

Domain Adaptation: the Key Enabler of Neural Network Equalizers in Coherent Optical Systems