Abstract
Statistical models such as linear regression drive numerous applications in computer vision and machine learning. The landscape of practical deployments of these formulations is dominated by forward regression models that estimate the parameters of a function mapping a set of p covariates, x, to a response variable, y. The less known alternative, Inverse Regression, offers various benefits that are much less explored in vision problems. The goal of this paper is to show how Inverse Regression in the “abundant” feature setting (i.e., many subsets of features are associated with the target label or response, as is the case for images), together with a statistical construction called Sufficient Reduction, yields highly flexible models that are a natural fit for model estimation tasks in vision. Specifically, we obtain formulations that provide relevance of individual covariates used in prediction, at the level of specific examples/samples — in a sense, explaining why a particular prediction was made. With no compromise in performance relative to other methods, an ability to interpret why a learning algorithm is behaving in a specific way for each prediction, adds significant value in numerous applications. We illustrate these properties and the benefits of Abundant Inverse Regression on three distinct applications.
| Original language | English |
|---|---|
| Title of host publication | Computer Vision - 14th European Conference, ECCV 2016, Proceedings |
| Editors | Jiri Matas, Nicu Sebe, Max Welling, Bastian Leibe |
| Publisher | Springer Verlag |
| Pages | 570-584 |
| Number of pages | 15 |
| ISBN (Print) | 9783319464862 |
| DOIs | |
| Publication status | Published - 2016 Jan 1 |
| Externally published | Yes |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 9907 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Fingerprint
Keywords
- Abundant regression
- Age estimation
- Alzheimer’s disease
- Inverse regression
- Kernel regression
- Temperature prediction
ASJC Scopus subject areas
- Theoretical Computer Science
- Computer Science(all)
Cite this
Abundant inverse regression using sufficient reduction and its applications. / Kim, Hyun Woo; Smith, Brandon M.; Adluru, Nagesh; Dyer, Charles R.; Johnson, Sterling C.; Singh, Vikas.
Computer Vision - 14th European Conference, ECCV 2016, Proceedings. ed. / Jiri Matas; Nicu Sebe; Max Welling; Bastian Leibe. Springer Verlag, 2016. p. 570-584 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9907 LNCS).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Abundant inverse regression using sufficient reduction and its applications
AU - Kim, Hyun Woo
AU - Smith, Brandon M.
AU - Adluru, Nagesh
AU - Dyer, Charles R.
AU - Johnson, Sterling C.
AU - Singh, Vikas
PY - 2016/1/1
Y1 - 2016/1/1
N2 - Statistical models such as linear regression drive numerous applications in computer vision and machine learning. The landscape of practical deployments of these formulations is dominated by forward regression models that estimate the parameters of a function mapping a set of p covariates, x, to a response variable, y. The less known alternative, Inverse Regression, offers various benefits that are much less explored in vision problems. The goal of this paper is to show how Inverse Regression in the “abundant” feature setting (i.e., many subsets of features are associated with the target label or response, as is the case for images), together with a statistical construction called Sufficient Reduction, yields highly flexible models that are a natural fit for model estimation tasks in vision. Specifically, we obtain formulations that provide relevance of individual covariates used in prediction, at the level of specific examples/samples — in a sense, explaining why a particular prediction was made. With no compromise in performance relative to other methods, an ability to interpret why a learning algorithm is behaving in a specific way for each prediction, adds significant value in numerous applications. We illustrate these properties and the benefits of Abundant Inverse Regression on three distinct applications.
AB - Statistical models such as linear regression drive numerous applications in computer vision and machine learning. The landscape of practical deployments of these formulations is dominated by forward regression models that estimate the parameters of a function mapping a set of p covariates, x, to a response variable, y. The less known alternative, Inverse Regression, offers various benefits that are much less explored in vision problems. The goal of this paper is to show how Inverse Regression in the “abundant” feature setting (i.e., many subsets of features are associated with the target label or response, as is the case for images), together with a statistical construction called Sufficient Reduction, yields highly flexible models that are a natural fit for model estimation tasks in vision. Specifically, we obtain formulations that provide relevance of individual covariates used in prediction, at the level of specific examples/samples — in a sense, explaining why a particular prediction was made. With no compromise in performance relative to other methods, an ability to interpret why a learning algorithm is behaving in a specific way for each prediction, adds significant value in numerous applications. We illustrate these properties and the benefits of Abundant Inverse Regression on three distinct applications.
KW - Abundant regression
KW - Age estimation
KW - Alzheimer’s disease
KW - Inverse regression
KW - Kernel regression
KW - Temperature prediction
UR - http://www.scopus.com/inward/record.url?scp=84990026591&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84990026591&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-46487-9_35
DO - 10.1007/978-3-319-46487-9_35
M3 - Conference contribution
AN - SCOPUS:84990026591
SN - 9783319464862
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 570
EP - 584
BT - Computer Vision - 14th European Conference, ECCV 2016, Proceedings
A2 - Matas, Jiri
A2 - Sebe, Nicu
A2 - Welling, Max
A2 - Leibe, Bastian
PB - Springer Verlag
ER -