Papers
7
Total Citations
199
H-Index
6
About
Scott Sanner is a leading researcher at the intersection of artificial intelligence, robotics, and decision-making under uncertainty. His work spans probabilistic reasoning, reinforcement learning, and autonomous systems, with a particular focus on enabling intelligent agents to operate effectively in complex, dynamic environments. Sanner made early contributions to mobile robotics with his work on object mapping in non-stationary environments (93 citations), introducing an expectation-maximization approach for detecting and adapting to environmental changes. He has since advanced the field of Bayesian optimisation for spatial-temporal monitoring (53 citations), developing sequential methods for costly-to-evaluate objective functions. A major theoretical contribution is his work on Symbolic Variable Elimination (SVE) for discrete and continuous graphical models (30 citations), providing a novel exact inference method for non-Gaussian distributions. Sanner has also explored risk-aware reinforcement learning through successor features and loss-calibrated Monte Carlo action selection, addressing critical challenges in sample efficiency and safe decision-making. His research continues to influence both theoretical foundations and practical applications in AI and robotics.
Research Focus
Key Achievements
Top Papers
- 1Towards object mapping in non-stationary environments with mobile robots93 citations · 2003
- 2Sequential Bayesian optimisation for spatial-temporal monitoring53 citations · 2014
- 3Symbolic Variable Elimination for Discrete and Continuous Graphical Models30 citations · 2021
- 4Risk-Aware Transfer in Reinforcement Learning using Successor Features9 citations · 2021
- 5Loss-Calibrated Monte Carlo Action Selection6 citations · 2015
- 6Sparse Kernel-SARSA(λ) with an Eligibility Trace6 citations · 2011
- 7How to spice up your planning under uncertainty research life2 citations · 2008