Aske Plaat
Papers
12
Total Citations
223
H-Index
9
About
Aske Plaat is a prominent computer scientist whose research sits at the intersection of reinforcement learning, large language models, and artificial intelligence. Based at Leiden University, Plaat has established himself as a leading voice in synthesizing and advancing complex AI methodologies, with his survey papers serving as essential reference points for researchers navigating rapidly evolving fields. His most influential work, "Deep Reinforcement Learning" (2022, 57 citations), offers a comprehensive treatment of the field, while his high-accuracy model-based reinforcement learning survey (2023, 40 citations) addresses the critical challenge of sample efficiency in deep RL systems — a persistent bottleneck for real-world deployment. His earlier contribution extending AlphaZero to continuous action spaces (2018) demonstrated creative problem-solving in bridging game-playing AI with practical robotics applications. More recently, Plaat has pivoted toward large language models, producing influential surveys on agentic LLMs and multi-step reasoning, reflecting the field's seismic shift. His work on affective computing with ChatGPT further reveals breadth across cognitive and emotional AI dimensions. With growing interest in human-robot co-creativity for older adults, Plaat's research increasingly connects technical innovation with meaningful societal impact, making his portfolio particularly valuable for students exploring both theoretical and applied AI frontiers.
Research Focus
Key Achievements
Top Papers
- 1Deep Reinforcement Learning57 citations · 2022
- 2High-accuracy model-based reinforcement learning, a survey40 citations · 2023
- 3Agentic Large Language Models, a Survey24 citations · 2025
- 4
- 5A0C: Alpha Zero in Continuous Action Space20 citations · 2018
- 6Multi-Step Reasoning with Large Language Models, a Survey13 citations · 2025
- 7Multi-Step Reasoning with Large Language Models, a Survey13 citations · 2024
- 8
- 9
- 10