In this talk, we delve into two pivotal subjects. The first topic revolves   around the development of hybrid graph models tailored to the complexities of   multi-modal data. We present a novel semi-supervised hypergraph learning   framework, specifically designed for diagnostic purposes. Our approach adopts   a hybrid perspective, where we introduce a new methodology centered on a dual   embedding strategy and a semi-explicit flow. To illustrate the efficacy of   our proposed model, we employ it within the realm of Alzheimer's disease   diagnosis, demonstrating its capacity to uncover latent relationships within   intricate multi-modal data.

Transitioning seamlessly to the second subject, we delve   into implicit neural representations. We introduce an innovative function   designed to harness the strengths of Strong Spatial and Frequency attributes,   marking a departure from conventional methods. Remarkably, our novel   technique showcases exceptional enhancements in performance across a diverse   array of downstream tasks, notably encompassing CT reconstruction and   denoising applications. Through rigorous experimentation, we elucidate the   advantages enabled by our approach.

Concluding our discourse, we shed light on emerging   trends within Diffusion models, reflecting on their evolving role and   potential. By amalgamating insights from hybrid graph models, implicit neural   representations, and diffusion modeling, this talk serves as a comprehensive   exploration of transformative techniques and trends in the field.

Biography:

Dr. Angelica is a Senior Research Associate at the Department   of Applied Mathematics and Theoretical Physics (DAMTP), University of   Cambridge. Her work focuses at the confluence of computational mathematics, computer   vision, and machine learning, where she addresses complex real-world problems   on a large scale.

Her expertise lies in developing   large-scale mathematical and machine learning models with minimal or even no   supervision. This has led to her being sought after for consultations by   various centers and companies.

She has large experience in   organising scientific meetings including BMVC2022 & BMVC 2023   (co-organiser), MIUA22 (co-organiser),  MICCAI Tutorials, ACCV   Tutorials, IGARSS Tutorials,  and GeoMedIA Workshop.

She serves as a current SIAM   SIAG/IS Officer.

For more info: https://angelicaiaviles.wordpress.com/