Learning to round for discrete labeling problems
Discrete labeling problems are often solved by formulating them as an integer program, and relaxing the integrality constraint to a continuous domain. While the continuous relaxation is closely related to the original integer program, its optimal solution is often fractional. Thus, the success of a...
Main Authors: | Mohapatra, P, Jawahar, C, Mudigonda, P |
---|---|
Format: | Conference item |
Published: |
PMLR
2018
|
Similar Items
-
Learning human poses from actions
by: Arun, A, et al.
Published: (2018) -
Parsimonious labeling
by: Dokania, P, et al.
Published: (2016) -
Efficient optimization for rank-based loss functions
by: Mohapatra, P, et al.
Published: (2018) -
Weakly supervised instance segmentation by learning annotation consistent instances
by: Arun, A, et al.
Published: (2020) -
Rounding-based moves for semi-metric labeling
by: Kumar, M, et al.
Published: (2016)