# Learning More Universal Representations for Transfer-Learning

arXiv: https://arxiv.org/abs/1712.09708 · HTML: https://mlsanity.com/paper/1712.09708/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Dec 2017.

## Built on by (most cited first, 6 of 6)

- [An Analysis of Initial Training Strategies for Exemplar-Free Class-Incremental Learning](https://mlsanity.com/paper/2308.11677/) (2023)
- [Deep Miner: A Deep and Multi-branch Network which Mines Rich and Diverse Features for Person Re-identification](https://mlsanity.com/paper/2102.09321/) (2021)
- [Neural Supervised Domain Adaptation by Augmenting Pre-trained Models with Random Units](https://mlsanity.com/paper/2106.04935/) (2021)
- [Learning Semantic Ambiguities for Zero-Shot Learning](https://mlsanity.com/paper/2201.01823/) (2022)
- [SCIDA: Self-Correction Integrated Domain Adaptation from Single- to Multi-label Aerial Images](https://mlsanity.com/paper/2108.06810/) (2021)
- [Novel Deep Neural Network Classifier Characterization Metrics with Applications to Dataless Evaluation](https://mlsanity.com/paper/2407.13000/) (2024)

## Abstract (arXiv)

A representation is supposed universal if it encodes any element of the visual world (e.g., objects, scenes) in any configuration (e.g., scale, context). While not expecting pure universal representations, the goal in the literature is to improve the universality level, starting from a representation with a certain level. To do so, the state-of-the-art consists in learning CNN-based representations on a diversified training problem (e.g., ImageNet modified by adding annotated data). While it effectively increases universality, such approach still requires a large amount of efforts to satisfy the needs in annotated data. In this work, we propose two methods to improve universality, but pay special attention to limit the need of annotated data. We also propose a unified framework of the methods based on the diversifying of the training problem. Finally, to better match Atkinson's cognitive study about universal human representations, we proposed to rely on the transfer-learning scheme as well as a new metric to evaluate universality. This latter, aims us to demonstrates the interest of our methods on 10 target-problems, relating to the classification task and a variety of visual domains.