Efficient Algorithms for Learning from Coarse Labels

08/22/2021
by   Dimitris Fotakis, et al.
0

For many learning problems one may not have access to fine grained label information; e.g., an image can be labeled as husky, dog, or even animal depending on the expertise of the annotator. In this work, we formalize these settings and study the problem of learning from such coarse data. Instead of observing the actual labels from a set 𝒵, we observe coarse labels corresponding to a partition of 𝒵 (or a mixture of partitions). Our main algorithmic result is that essentially any problem learnable from fine grained labels can also be learned efficiently when the coarse data are sufficiently informative. We obtain our result through a generic reduction for answering Statistical Queries (SQ) over fine grained labels given only coarse labels. The number of coarse labels required depends polynomially on the information distortion due to coarsening and the number of fine labels |𝒵|. We also investigate the case of (infinitely many) real valued labels focusing on a central problem in censored and truncated statistics: Gaussian mean estimation from coarse data. We provide an efficient algorithm when the sets in the partition are convex and establish that the problem is NP-hard even for very simple non-convex sets.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/25/2020

Grafit: Learning fine-grained image representations with coarse labels

This paper tackles the problem of learning a finer representation than t...
research
08/16/2022

Reinforcement Learning to Rank with Coarse-grained Labels

Ranking lies at the core of many Information Retrieval (IR) tasks. While...
research
12/19/2020

Towards Coarse and Fine-grained Multi-Graph Multi-Label Learning

Multi-graph multi-label learning (Mgml) is a supervised learning framewo...
research
11/05/2018

Improving Span-based Question Answering Systems with Coarsely Labeled Data

We study approaches to improve fine-grained short answer Question Answer...
research
11/18/2020

Extracting and Learning Fine-Grained Labels from Chest Radiographs

Chest radiographs are the most common diagnostic exam in emergency rooms...
research
12/04/2020

Rethinking movie genre classification with fine-grained semantic clustering

Movie genre classification is an active research area in machine learnin...
research
09/14/2023

Methodologies for Selection of Optimal Sites for Renewable Energy Under a Diverse Set of Constraints and Objectives

In this paper, we present methodologies for optimal selection for renewa...

Please sign up or login with your details

Forgot password? Click here to reset