Auto Seg-Loss: Searching Metric Surrogates for Semantic Segmentation

10/15/2020
by   Hao Li, et al.
10

We propose a general framework for searching surrogate losses for mainstream semantic segmentation metrics. This is in contrast to existing loss functions manually designed for individual metrics. The searched surrogate losses can generalize well to other datasets and networks. Extensive experiments on PASCAL VOC and Cityscapes demonstrate the effectiveness of our approach. Code shall be released.

READ FULL TEXT

page 13

page 14

research
03/25/2021

AutoLoss-Zero: Searching Loss Functions from Scratch for Generic Tasks

Significant progress has been achieved in automating the design of vario...
research
05/24/2019

Learning Surrogate Losses

The minimization of loss functions is the heart and soul of Machine Lear...
research
05/24/2017

Optimization of the Jaccard index for image segmentation with the Lovász hinge

The Jaccard loss, commonly referred to as the intersection-over-union lo...
research
07/17/2019

An Embedding Framework for Consistent Polyhedral Surrogates

We formalize and study the natural approach of designing convex surrogat...
research
12/28/2016

The Pessimistic Limits of Margin-based Losses in Semi-supervised Learning

We show that for linear classifiers defined by convex margin-based surro...
research
11/09/2021

Tightening the Approximation Error of Adversarial Risk with Auto Loss Function Search

Numerous studies have demonstrated that deep neural networks are easily ...
research
02/26/2022

Relational Surrogate Loss Learning

Evaluation metrics in machine learning are often hardly taken as loss fu...

Please sign up or login with your details

Forgot password? Click here to reset