Ruben Glatt

About

Ruben Glatt is a researcher specializing in multiagent reinforcement learning (MARL), with a particular focus on accelerating learning efficiency in complex, multi-agent environments. His work addresses one of the fundamental challenges in reinforcement learning: the prohibitively large number of environment interactions typically required for agents to develop effective policies — a problem that becomes significantly more acute when multiple autonomous agents must learn simultaneously. Glatt's most influential contribution, "Simultaneously Learning and Advising in Multiagent Reinforcement Learning" (2017), has garnered over 104 citations and introduces an innovative framework in which agents can act as both learners and advisors concurrently. By drawing on teacher-student learning paradigms, his research enables agents to share knowledge dynamically, substantially reducing the time and data required to converge on optimal behavior. This work represents a meaningful step forward in making reinforcement learning more practical for real-world, multi-agent deployments. Through his advising framework research, Glatt has helped establish a foundational approach to knowledge transfer in MARL systems. His contributions are particularly relevant to researchers working on cooperative AI, autonomous systems, and scalable machine learning, making his work essential reading for anyone exploring efficient multi-agent coordination strategies.

Research Focus

Key Achievements

Top Papers

1. 1
   Simultaneously Learning and Advising in Multiagent Reinforcement Learning

   104 citations · 2017
2. 2
   An Advising Framework for Multiagent Reinforcement Learning Systems

   5 citations · 2017
3. 3
   Simultaneously Learning and Advising in Multiagent Reinforcement Learning

   2 citations · 2017

Key Collaborators

Contact & Links