Haphazard Inputs as Images in Online Learning

2025-04-03

AI Summary

In online learning, input feature dimensions vary dynamically (“haphazard inputs”), rendering mainstream vision models—such as ResNet and ViT—unsuitable for direct application due to their reliance on fixed-dimensional inputs. To address this, we propose the first model-agnostic online image representation framework that maps variable-length sequences into fixed-size 2D images in real time. Our method employs dynamic sequence padding and optimized spatial layout to achieve structured encoding, followed by integration with standard vision backbones for end-to-end learning. This eliminates the need for architecture-specific designs, enabling plug-and-play deployment of state-of-the-art vision models in online learning settings. Extensive evaluation across four public benchmarks demonstrates significant improvements in model robustness, generalization, and scalability. The implementation is publicly available.

Technology Category

• Computer Vision: Large Vision Models
• Machine Learning: Representation Learning
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

The field of varying feature space in online learning settings, also known as haphazard inputs, is very prominent nowadays due to its applicability in various fields. However, the current solutions to haphazard inputs are model-dependent and cannot benefit from the existing advanced deep-learning methods, which necessitate inputs of fixed dimensions. Therefore, we propose to transform the varying feature space in an online learning setting to a fixed-dimension image representation on the fly. This simple yet novel approach is model-agnostic, allowing any vision-based models to be applicable for haphazard inputs, as demonstrated using ResNet and ViT. The image representation handles the inconsistent input data seamlessly, making our proposed approach scalable and robust. We show the efficacy of our method on four publicly available datasets. The code is available at https://github.com/Rohit102497/HaphazardInputsAsImages.

Problem

• Transforming varying feature space to fixed-dimension images
• Enabling vision models for haphazard inputs
• Handling inconsistent data in online learning

Innovation

• Transform varying feature space to fixed-dimension images
• Model-agnostic approach for vision-based models
• Seamlessly handles inconsistent input data robustly