Rokas Bendikas

Rokas is currently pursuing a PhD in Foundational AI at University College London and the University of California, San Diego under the supervision of Prof. Danail Stoyanov and Prof. Hao Su. His research is focused on model-based Reinforcement Learning and 3D scene understanding in robot control domains. Rokas is dedicated to empowering robots with the ability to learn intelligent behaviours through the spatial understanding of their surroundings.

During his Master’s studies at Imperial College London, Rokas completed his thesis at the Dyson Robotics Lab, under the supervision of Prof. Andrew Davison. His project focused on developing an imagination-augmented DQN algorithm, enabling robots to learn optimal object manipulation strategies in dynamic environments by “imagining” the future.

In his undergraduate years in Biomedical Engineering at King’s College London, Rokas contributed as a research assistant in the Cardio-Electro Magnetic Research Group (CEMRG). The group, led by Dr. Caroline Roney and Prof. Steven Niederer, researches the mechanisms of atrial fibrillation using Deep Learning and Computational Modelling.

Rokas also completed a software engineering internship at MathWorks and a research internship at Qualcomm.

• Robot Control
• Reinforcement Learning
• 3D Scene Understanding

• Reinforcement Learning
• Robotic Control Theory and Systems