IDEAL : Interpretable-by-Design ALgorithms for learning from foundation feature spaces

Angelov, P. and Kangin, D. and Zhang, Z. (2025) IDEAL : Interpretable-by-Design ALgorithms for learning from foundation feature spaces. Neurocomputing, 626: 129464. ISSN 0925-2312

Text (revision_IDEAL)
revision_IDEAL.pdf - Accepted Version
Available under License Creative Commons Attribution.
Download (13MB)

Abstract

The advance of foundation models (FM) makes it possible to avoid parametric tuning for transfer learning, taking advantage of pretrained feature spaces. In this study, we define a framework called IDEAL (Interpretable-by-design DEep learning ALgorithms) which tackles the problem of interpretable transfer learning by recasting the standard supervised classification problem into a function of similarity to a set of prototypes derived from the training data. This framework generalises previously-known prototypical approaches, such as ProtoPNet, xDNN and DNC, and decomposes the overall problem into two inherently connected stages: (A) feature extraction (FE), which maps the raw features of real-world data into a latent space, and (B) identification of representative prototypes and decision making based on similarity and association between the query and the prototypes. This addresses the issue of interpretability (stage B) while retaining the benefits of pretrained deep learning (DL) models. On a range of datasets (CIFAR-10, CIFAR-100, CalTech101, STL-10, Oxford-IIIT Pet, EuroSAT), we demonstrate, through an extensive set of experiments, how the choice of the latent space, prototype selection, and finetuning of the latent space affect accuracy and generalisation of the models on transfer learning scenarios for different backbones. Building upon this knowledge, we demonstrate that the proposed framework helps achieve an advantage over state-of-the-art baselines in class-incremental learning. The key findings can be summarised as follows: (1) the setting allows interpretability through prototypes, (2) lack of finetuning helps circumvent the issue of catastrophic forgetting, allowing efficient class-incremental transfer learning, while mitigating the issue of confounding bias, and (3) ViT architectures narrow the gap between finetuned and non-finetuned models allowing for transfer learning in a fraction of time without finetuning of the feature space on a target dataset with iterative supervised methods.