Eleftherios Triantafyllidis

My name is Eleftherios, a recent PhD graduate at the University of Edinburgh. My research lies at the intriguing intersection of Machine Learning, Robotics, and Human Factors.

My doctoral work primarily revolves around developing innovative, applied ML algorithms for robotics, drawing inspiration from biological perspectives [1]. Fascinated by the advanced cognitive capabilities of the human brain, I initially aimed to delve into the workings of the human sensory system and its perception of its surroundings. To this endeavour, I focused extensively on how different sensory feedback states when utilising multimodal interfaces, contribute towards task performance during motor movements, reflected on increasingly more complex spatial settings related to manipulation [2]. Subsequently, I sought to measure and quantify these behaviours, leading me to devise a novel metric capable of gauging and quantifying human motor performance in high-dimensional space [3,4,5]. Building upon my earlier work, my penultimate PhD aim was to explore effective methods for transferring human capabilities to embodied intelligent robotic agents. Inspired by a biological standpoint, I developed and published a new Hybrid Hierarchical Learning (HHL) framework, known as the RObotic MAnipulation Network (ROMAN), to address the challenges of significantly complex, long-horizon sequential manipulation tasks in robotics [6]. To further solidify this gap, my last PhD aim harnesses the power of language for exploratory potential. To this end, I introduced the Intrinsically Guided Exploration from Large Language Models (IGE-LLMs) framework, tailored to address the challenges of notably intricate, long-horizon of sparse rewards tasks [7].

News Doctoral Thesis My doctoral thesis Advancements in Sensory-Motor Perception and Biologically-Inspired Hierarchical Learning for Embodied Intelligence was succesfully defended without corrections!
News Research (Conference) Our latest paper Intrinsic Language-Guided Exploration for Complex Long-Horizon Robotic Manipulation Tasks has been accepted at ICRA 2024 (Pre-Print Available)!
News Research (Journal) Our paper Hybrid Hierarchical Learning for Solving Complex Sequential Tasks Using the RObotic MAnipulation Network - ROMAN is now available in Nature Machine Intelligence (2023)!
News Research (Journal) Our paper Identifying Important Sensory Feedback for Learning Locomotion Skills is now available in Nature Machine Intelligence (2023) and also made the cover!
News Research (Journal) Our paper Modular Neural Network Policies for Learning In-Flight Object Catching with a Robot Hand-Arm System was accepted in IEEE IROS 2023!
News ICRA 2022 Invitation! Our paper Metrics for 3D Object Pointing and Manipulation in Virtual Reality was invited and presented in ICRA 2022! Poster.
News Research (Journal) Our paper Metrics for 3D Object Pointing and Manipulation in Virtual Reality: The Introduction and Validation of a Novel Approach in Measuring Human Performance was accepted at the IEEE Robotics and Automation Magazine (RAM) 2021!
News Research (Conference) Our paper The Challenges in Modeling Human Performance in 3D Space with Fitts’ Law was accepted at the ACM Conference on Human Factors in Computing Systems (CHI) 2021!
News Research (Conference) Our paper Considerations and Challenges of Measuring Operator Performance in Telepresence and Teleoperation Entailing Mixed Reality Technologies was accepted at the ACM Conference on Human Factors in Computing Systems Workshop - Evaluating User Experiences in Mixed Reality (CHI) 2021!
News Best Student Case Study of 2021! Our paper How Well Am I Doing?: Measuring Human Performance in Teleoperation – The Introduction of a Novel Approach was selected as the best case study in the Annual Review of the CDT - RAS!
News Research (Journal) Our paper Study of Multimodal Interfaces and the Improvements on Teleoperation was accepted at the IEEE Access Journal 2020!
News Research (Book Chapter) Our submission Robot Intelligence for Real-World Applications was accepted and included as a book chapter in the IET - AI for Emerging Verticals: Human-Robot Computing, Sensing and Networking, 2020!