Maria Tzelepi

Papers

Total Citations

H-Index

About

Maria Tzelepi is a leading researcher at the intersection of deep learning and robotics, with a primary focus on creating efficient, lightweight neural architectures. Her most impactful contribution is the development of **OpenDR**, an open-source toolkit designed to bridge the critical gap between standard deep learning frameworks and the specific demands of robotics. This work, which has garnered 24 citations since 2022, enables high-performance, low-footprint deep learning for embodied systems, addressing the steep learning curve and methodological differences that have hindered robotics adoption. Tzelepi has also advanced the field of model compression through her work on **Multilayer Online Self-Acquired Knowledge Distillation**, a novel method that circumvents the computational and memory burdens of traditional offline distillation by allowing a network to learn from its own intermediate representations during training. Her research on lightweight deep learning further underscores her commitment to deploying sophisticated AI on resource-constrained robotic platforms. By tackling both the software infrastructure and the algorithmic efficiency of deep learning, Tzelepi is paving the way for more accessible, real-time, and autonomous robotic systems.