Dissertation Defence: Compositional and Adaptive Visual Recognition Under Domain Shift

Ahmed Radwan, supervised by Dr. Mohamed S. Shehata, will defend their dissertation titled “Compositional and Adaptive Visual Recognition Under Domain Shift” in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Computer Science.

An abstract for Ahmed Radwan’s dissertation is included below.

Examinations are open to all members of the campus community as well as the general public. This examination will be offered in hybrid format.  Registration is not required to attend in person, but please email mohamed.sami.shehata@ubc.ca to receive the Zoom link for this exam.

Abstract

Human vision is remarkably robust, data-efficient, and adaptive. People can recognize objects across changes in style, viewpoint, lighting, modality, and context, often from only a few examples. Modern deep learning systems, despite their impressive performance, still depend heavily on large labelled datasets and often degrade when the test distribution differs from the training distribution. This brittleness is especially consequential in settings where labels are scarce, data are decentralized, or privacy constraints prevent the construction of large centralized datasets.

This thesis argues that robust visual recognition requires moving beyond the dominant view of images as global patterns mapped directly to labels. Instead, it develops the view that recognition should be grounded in the structure of visual evidence, since objects and scenes are composed of parts, regions, relations, and context-dependent cues, not all of which are equally relevant to the recognition decision. Inspired by the robustness of biological perception, the thesis organizes this view around three computational principles. First, visual representations should expose compositional structure rather than collapse images into monolithic embeddings. Second, recognition should selectively compare the evidence that is shared, discriminative, and stable while suppressing clutter, background, and nuisance variation. Third, inference should adapt to the input, domain, or task without requiring full retraining or access to target-domain supervision.

The contributions of this thesis instantiate these principles across several recognition settings. It develops architectures and learning strategies that encourage part–whole and region-level structure; training-free and few-shot inference methods that compare images through selected local evidence rather than global similarity alone; adaptive specialization mechanisms that preserve pretrained knowledge while changing the effective computation for new domains; and robustness-oriented optimization strategies that reduce reliance on nuisance-sensitive cues. Although these methods differ in form, they share the common goal of making recognition depend less on superficial appearance correlations and more on structured, reusable, and task-relevant visual evidence. Together, these contributions present a computational pathway toward brain-inspired visual recognition under real-world constraints.

The thesis shows that compositional structure, selective evidence use, and adaptive inference provide a coherent foundation for recognition under domain shift, data scarcity, and privacy constraints. Across natural images, medical imaging, federated learning, cross-domain few-shot recognition, and video understanding, the work demonstrates that robust visual recognition is not only a matter of larger models or more data, but of representing and using visual evidence in a more structured and adaptive way.