Generative Reasoning Integrated Label Noise Robust Deep Image Representation Learning

TL;DR

The paper introduces GRID, a novel deep learning approach that combines generative and discriminative reasoning to improve image representation learning robustness against noisy labels, enhancing accuracy without depending on specific architectures.

Contribution

GRID uniquely integrates generative reasoning with discriminative learning via a supervised variational autoencoder to detect and mitigate label noise effects.

Findings

Outperforms state-of-the-art methods in noisy label scenarios.

Effectively detects noisy labels using generative reasoning.

Applicable across various IRL tasks without architecture constraints.

Abstract

The development of deep learning based image representation learning (IRL) methods has attracted great attention for various image understanding problems. Most of these methods require the availability of a high quantity and quality of annotated training images, which can be time-consuming and costly to gather. To reduce labeling costs, crowdsourced data, automatic labeling procedures or citizen science projects can be considered. However, such approaches increase the risk of including label noise in training data. It may result in overfitting on noisy labels when discriminative reasoning is employed. This leads to sub-optimal learning procedures, and thus inaccurate characterization of images. To address this, we introduce a generative reasoning integrated label noise robust deep representation learning (GRID) approach. Our approach aims to model the complementary characteristics of discriminative and generative reasoning for IRL under noisy labels. To this end, we first integrate generative reasoning into discriminative reasoning through a supervised variational autoencoder. This allows GRID to automatically detect training samples with noisy labels. Then, through our label noise robust hybrid representation learning strategy, GRID adjusts the whole learning procedure for IRL of these samples through generative reasoning and that of other samples through discriminative reasoning. Our approach learns discriminative image representations while preventing interference of noisy labels independently from the IRL method being selected. Thus, unlike the existing methods, GRID does not depend on the type of annotation, neural network architecture, loss function or learning task, and thus can be directly utilized for various problems. Experimental results show its effectiveness compared to state-of-the-art methods. The code of GRID is publicly available at https://github.com/gencersumbul/GRID.