Deep Imbalanced Regression: Challenges, Methods, and Applications
Name
Zha-kzha-SM-EECS-2022-thesis.pdf
Description
Thesis PDF
Size
1.29 MB
Format
Adobe PDF
Checksum (MD5)
790bf2b84e52e62318647921f109ba93
Author(s)
Zha, Kaiwen
Advisor(s)
Katabi, Dina
Date Issued
May 2022
Publisher
Massachusetts Institute of Technology
Abstract
Real-world data often exhibit imbalanced distributions, where certain target values have significantly fewer observations. Existing techniques for dealing with imbalanced data focus on targets with categorical indices, i.e., different classes. However, many tasks involve continuous targets, where hard boundaries between classes do not exist. We define Deep Imbalanced Regression (DIR) as learning from such imbalanced data with continuous targets, dealing with potential missing data for certain target values, and generalizing to the entire target range. Motivated by the intrinsic difference between categorical and continuous label space, we propose distribution smoothing for both labels and features, which explicitly acknowledges the effects of nearby targets, and calibrates both label and learned feature distributions. We curate and benchmark large-scale DIR datasets from common real-world tasks in computer vision, natural language processing, and healthcare domains. Extensive experiments verify the superior performance of our strategies. Our work fills the gap in benchmarks and techniques for practical imbalanced regression problems.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
In Copyright - Educational Use Permitted
Copyright MIT
Persistent DSpace Link
