Papers
17
Total Citations
1,203
H-Index
12
About
Michael L. Littman is a prominent computer scientist whose research spans reinforcement learning, human-robot interaction, and planning under uncertainty. His foundational work on partially observable environments — cited over 660 times — helped establish scalable frameworks for learning in real-world settings where agents lack complete state information, a cornerstone contribution to the field of sequential decision-making. Littman has made significant strides in bridging machine learning and human collaboration. His investigations into interactive reinforcement learning explore how robots and agents can effectively learn from non-expert human feedback — including evaluative signals, natural language commands, and even implicit behavioral cues — making intelligent systems more accessible to everyday users. His 2015 work grounding English commands to reward functions exemplifies this commitment to intuitive human-machine communication. Beyond human feedback, Littman has tackled challenges in continuous state spaces, hierarchical planning with abstract Markov Decision Processes, and model-based exploration, demonstrating remarkable breadth. His research consistently addresses practical deployment concerns, such as adapting agent action speed to improve learning outcomes with non-expert trainers. Collectively, his body of work — spanning decades and accumulating over a thousand citations — has meaningfully shaped how intelligent agents learn, plan, and interact within complex, human-centered environments.
Research Focus
Key Achievements
Top Papers
- 1Learning policies for partially observable environments: Scaling up662 citations · 1995
- 2Interactive Learning from Policy-Dependent Human Feedback108 citations · 2017
- 3
- 4Grounding English Commands to Reward Functions65 citations · 2015
- 5Efficient reinforcement learning with relocatable action models64 citations · 2007
- 6Planning with Abstract Markov Decision Processes49 citations · 2017
- 7
- 8
- 9
- 10CORL: A Continuous-state Offset-dynamics Reinforcement Learner22 citations · 2012