Timezone: »
Transferability Between Regression Tasks
Cuong Ngoc Nguyen · Phong Tran The · Lam Ho · Vu Dinh · Anh Tran · Tal Hassner · Cuong V. Nguyen
Event URL: https://openreview.net/forum?id=2dCAMPKgCo »
Transfer learning has been a widely used technique to adapt a deep learning model trained for one task to another when there is a data distribution shift between these tasks. To improve the effectiveness of transfer learning and to understand relationships between tasks, we consider the problem of transferability estimation between regression tasks and propose two novel transferability estimators that are simple, computationally efficient, yet effective and theoretically grounded. We test our proposed methods extensively in various challenging, practical scenarios and show they significantly outperform existing state-of-the-art regression task transferability estimators in both accuracy and efficiency.
Author Information
Cuong Ngoc Nguyen (Florida International University)
Phong Tran The (MBZUAI)
Lam Ho (University of Dalhousie)
Vu Dinh (University of Delaware)
Anh Tran (VinAI Research)
Tal Hassner (Facebook)
Cuong V. Nguyen (Florida International University)
More from the Same Authors
-
2022 Poster: QC-StyleGAN - Quality Controllable Image Generation and Manipulation »
Dat Viet Thanh Nguyen · Phong Tran The · Tan M. Dinh · Cuong Pham · Anh Tran -
2021 Poster: Single Layer Predictive Normalized Maximum Likelihood for Out-of-Distribution Detection »
Koby Bibas · Meir Feder · Tal Hassner -
2021 Poster: Exploiting Domain-Specific Features to Enhance Domain Generalization »
Manh-Ha Bui · Toan Tran · Anh Tran · Dinh Phung -
2021 Poster: On Learning Domain-Invariant Representations for Transfer Learning with Multiple Sources »
Trung Phung · Trung Le · Tung-Long Vuong · Toan Tran · Anh Tran · Hung Bui · Dinh Phung -
2020 Poster: Consistent feature selection for analytic deep neural networks »
Vu Dinh · Lam Ho -
2020 Poster: Input-Aware Dynamic Backdoor Attack »
Tuan Anh Nguyen · Anh Tran -
2016 Poster: Adaptive Maximization of Pointwise Submodular Functions With Budget Constraint »
Nguyen Viet Cuong · Huan Xu -
2013 Poster: Active Learning for Probabilistic Hypotheses Using the Maximum Gibbs Error Criterion »
Nguyen Viet Cuong · Wee Sun Lee · Nan Ye · Kian Ming Adam Chai · Hai Leong Chieu