Robust Transfer Learning

Robust transfer learning aims to improve the performance of machine learning models on new tasks by leveraging knowledge from previously trained models, while mitigating the negative effects of differences in data distributions between source and target domains. Current research focuses on developing methods to handle unreliable source data, covariate shifts, and model discrepancies, often employing techniques like fused regularization, thresholding to balance source and target performance, and adaptive pipelines that select appropriate algorithms. These advancements are crucial for improving the reliability and efficiency of machine learning across diverse applications, particularly in areas like computer vision (e.g., pothole detection) and agriculture (e.g., pest identification), where data scarcity or variability is common.

Papers

October 8, 2024

Robust Transfer Learning for Active Level Set Estimation with Locally Adaptive Gaussian Process Prior
Giang Ngo, Dang Nguyen, Sunil Gupta
Transfer Learning Gaussian Process Estimation Task Prior Knowledge Black Box Function Level Set Estimation Robust Transfer Learning

April 1, 2024

TransFusion: Covariate-Shift Robust Transfer Learning for High-Dimensional Regression
Zelin He, Ying Sun, Jingyuan Liu, Runze Li
High Dimensional Distribution Shift Covariate Shift Blood Transfusion Model Shift Robust Transfer Learning

October 6, 2023

Robust Transfer Learning with Unreliable Source Data
Jianqing Fan, Cheng Gao, Jason M. Klusowski
Transfer Learning Unreliable Source Bayes Classifier Weak Signal Robustness Transfer Robust Transfer Learning

August 7, 2023

RoadScan: A Novel and Robust Transfer Learning Framework for Autonomous Pothole Detection in Roads
Guruprasad Parasnis, Anmol Chokshi, Vansh Jain, Kailas Devadkar
Deep Learning Built Road Object Detection Algorithm Pothole Detection Robust Transfer Learning Pothole Encounter

May 28, 2023

JutePestDetect: An Intelligent Approach for Jute Pest Identification Using Fine-Tuned Transfer Learning
Md. Simul Hasan Talukder, Mohammad Raziuddin Chowdhury, Md Sakib Ullah Sourav, Abdullah Al Rakin, Shabbir Ahmed Shuvo, Rejwan Bin Sulaiman, Musarrat Saberin Nipun, Muntarin Islam, Mst Rumpa Islam, Md Aminul Islam, Zubaer Haque
Fine Tuning Pest Detection Pest Identification Robust Transfer Learning

April 30, 2023

The ART of Transfer Learning: An Adaptive and Robust Pipeline
Boxiang Wang, Yunan Wu, Chenglong Ye
Transfer Learning Adaptive Importance Learning Rate Adaptive Transfer Learning Robust Transfer Learning Generic Machine Learning

March 26, 2023

BlackVIP: Black-Box Visual Prompting for Robust Transfer Learning
Changdae Oh, Hyeji Hwang, Hee-young Lee, YongTaek Lim, Geunyoung Jung, Jiyoung Jung, Hosik Choi, Kyungwoo Song
Large Model Visual Prompting Large Scale Pre Trained Model Shot Adaptation Efficient Transfer Learning Robust Adaptation Robust Transfer Learning

July 12, 2022

A Data-Based Perspective on Transfer Learning
Saachi Jain, Hadi Salman, Alaa Khaddaj, Eric Wong, Sung Min Park, Aleksander Madry
Transfer Learning Supervised ImageNet Pre Training Data Source Data Perspective Robust Transfer Learning

January 20, 2022

An Empirical Investigation of Model-to-Model Distribution Shifts in Trained Convolutional Filters
Paul Gavrikov, Janis Keuper
Distribution Shift CNN Model Convolutional Filter Robust Transfer Learning

Robust Transfer Learning

Papers

Robust Transfer Learning for Active Level Set Estimation with Locally Adaptive Gaussian Process Prior

TransFusion: Covariate-Shift Robust Transfer Learning for High-Dimensional Regression

Robust Transfer Learning with Unreliable Source Data

RoadScan: A Novel and Robust Transfer Learning Framework for Autonomous Pothole Detection in Roads

JutePestDetect: An Intelligent Approach for Jute Pest Identification Using Fine-Tuned Transfer Learning

The ART of Transfer Learning: An Adaptive and Robust Pipeline

BlackVIP: Black-Box Visual Prompting for Robust Transfer Learning

A Data-Based Perspective on Transfer Learning

An Empirical Investigation of Model-to-Model Distribution Shifts in Trained Convolutional Filters