Papers
9
Total Citations
377
H-Index
6
About
Vassilis Vassiliades is a prominent researcher specializing in evolutionary robotics, reinforcement learning, and quality-diversity optimization. His work centers on enabling robots to adapt intelligently to real-world challenges, particularly hardware failures and data-scarce environments. Among his most impactful contributions is his research on reset-free trial-and-error learning, which allows robots to recover from damage through autonomous experimentation without human intervention — a critical capability for deployment in scenarios like post-disaster rescue. This work has accumulated over 105 citations, reflecting its significance to the robotics community. Complementing this, his highly cited work on black-box data-efficient policy search (93 citations) addresses one of robotics' most pressing challenges: teaching robots effective behaviors using only a handful of trials, a problem he further examined in a dedicated survey on policy search algorithms. Vassiliades has also made foundational contributions to quality-diversity optimization, a novel branch of stochastic optimization that seeks diverse, high-performing solutions simultaneously. His development of the elite hypervolume concept (67 citations) advanced MAP-Elites-style algorithms by leveraging correlations across behavioral niches. Collectively, his research has shaped modern approaches to resilient, efficient robot learning, earning him a strong international reputation in the field.
Research Focus
Key Achievements
Top Papers
- 1Reset-free Trial-and-Error Learning for Robot Damage Recovery105 citations · 2017
- 2Black-box data-efficient policy search for robotics93 citations · 2017
- 3Quality-Diversity Optimization: A Novel Branch of Stochastic Optimization77 citations · 2021
- 4Discovering the elite hypervolume by leveraging interspecies correlation67 citations · 2018
- 5
- 6
- 7
- 8
- 9Black-Box Data-efficient Policy Search for Robotics2 citations · 2017